Research Computing

Synthetic Network Data Generation for Analyst Training

2026-04-01T00:00:00Z

Synthetic Network Data Generation for Analyst Training Smith, Liam; Wright, Matthew Rapidly evolving cyber threats demand continuous, high-fidelity training for defense analysts. However, generating realistic network traffic datasets creates a significant barrier to entry, often requiring extensive virtualization infrastructure, specialized hardware, and knowledge in cyber range administration. This paper introduces a streamlined architecture, called Generative Packet Captures (GenCap), built upon the foundational capabilities of the FOSR benign traffic generator and the ID2T attack injector. By abstracting these complex tools behind an automated orchestration layer, it enables users to generate scenario-specific PCAP files on demand. This approach democratizes access to training data, allowing analysts to create rigorous network defense scenarios without the need for complex provisioning or systems engineering knowledge.

Office AI Automation using Existing DAF-Approved Software

2026-03-20T00:00:00Z

Office AI Automation using Existing DAF-Approved Software Cui, Wei; Kennedy, Laura The Department of the Air Force (DAF) continues to face mounting administrative workloads that hinder mission focus and operational efficiency. Executive officers and staff officers spend substantial time generating reports, managing emails, routing documents, and organizing taskers across multiple systems. This paper presents the Smart Executive Assistant, an office AI initiative to automate repetitive administrative tasks using existing DAF-approved technologies without a new Authority- To-Operate (ATO). By integrating DAF 365 applications, Power Automate, and approved large language models (LLMs) within secure IL5 and IL6 environments, this solution seeks to reduce time spent on low-value administrative processes by 90% while maintaining compliance and data security.

AI for Scalable Defensive Cyber Log Analysis

2026-03-20T00:00:00Z

AI for Scalable Defensive Cyber Log Analysis Schofield, Catherine; Jananthan, Hayden; Kepner, Jeremy Centralized cyber logging platforms ingest large volumes of heterogeneous telemetry, yet high dimensionality and query-driven workflows often limit scalable analytic insight on these systems. This work presents an automated pipeline for ingesting, characterizing, and analyzing large-scale hostbased logs using sparse representations and distribution-aware statistics. A systematic dimensional analysis reduces hundreds of raw log fields to a small set of informative dimensions suitable for aggregation across extended time windows. Temporal analysis of the reduced representation reveals coordinated deviations in activity volume and distributional behavior that are not apparent in individual log streams. The results demonstrate that dimensional reduction enables scalable, interpretable analysis of enterprise cyber telemetry. Furthermore, these results were obtained using host-based sensors designed for event-oriented point-defense and demonstrate the feasibility of integrating such sensors to enable long-range, long-duration area defense.

Cross-Aircraft Flight Phase Classification Using ADS-B Data and Transfer Learning

2026-03-20T00:00:00Z

Cross-Aircraft Flight Phase Classification Using ADS-B Data and Transfer Learning Kiefer, Jacob; Alemany, Sheila Flight phase identification (FPI) approaches that apply traditional machine learning techniques are expensive to scale, difficult to generalize across platforms, and frequently unavailable in permissive or distributed training environments. We propose a scalable, data-driven pipeline for automatic FPI using open-source Automatic Dependent Surveillance-Broadcast (ADS-B) data, with an emphasis on cross-aircraft generalization through transfer learning. Leveraging ADS-B telemetry from USAF Initial Flight Training aircraft, a neural network classifier is trained on Diamond DA-20 flight data and evaluated on Texan T-6 aircraft under zero-shot and fine-tuned transfer learning conditions. We describe a robust ADS-B preprocessing pipeline integrating digital elevation model (DEM) data, a data labeling strategy using unsupervised learning, and a transfer learning approach enabling adaptation across aircraft types with limited labeled data. Our results demonstrate that transfer learning significantly improves classification accuracy for flight phases with limited data, highlighting the potential of ADS-B-based models to support scalable, behavior-aware airspace intelligence across heterogeneous fleets and permissive environments. This research advances FPI capabilities for USAF training analysis and broader operational priorities in autonomy, situational awareness, and data-driven decision support.

Neural Networks for Stress Intensity Factor Vertex Prediction

2026-03-20T00:00:00Z

Neural Networks for Stress Intensity Factor Vertex Prediction Hokaj, Ian; Ghanem, Janelle Structural fatigue in aging metallic aircraft is a primary driver of sustainment costs for the U.S. Air Force, significantly impacting fleet readiness. Fatigue life prediction tools like AFGROW depend on interpolating between computationally expensive stress intensity factors (K-solutions) to approximate unknown values. However, interpolation errors in the current approach introduce uncertainty and force overly conservative maintenance schedules. This paper investigates the use of a machine learning surrogate to replace AFGROW’s dimensionreduction interpolation for the finite-width c orner-cracked hole geometry. We developed a robust data processing pipeline for a large FEA dataset and trained a neural network model. Our results reveal a critical insight: the surrogate model offers substantial performance gains over AFGROW’s interpolation in low-data regimes, emphasizing both the potential of the model and its sensitivity to dataset size. For the original, sparse dataset—which is characteristic of computationally expensive problems—the neural network significantly o utperformed the baseline interpolation, reducing the mean absolute percentage error (MAPE) by over 40% (from 2.77% to 1.60%) and achieving an R² value exceeding 0.99. However, experiments on synthetically generated dense datasets showed that the traditional interpolation method becomes more accurate as the data grid becomes less sparse. This study concludes that while neural network surrogates offer a powerful, high-fidelity solution for computationally intensive engineering problems, their adoption should be guided by a careful analysis of data density after dataset has been cleaned of outliers. It also highlights the necessity of employing rigorous, application-relevant validation strategies that move beyond simplistic random splits to accurately assess model performance.

Evaluating Adaptive AI for Contracting Officer Readiness: Design and Pedagogical Proposal for the Warrant Board RAG Chatbot

2026-03-20T00:00:00Z

Evaluating Adaptive AI for Contracting Officer Readiness: Design and Pedagogical Proposal for the Warrant Board RAG Chatbot Mullen, Julia; Grosvenor, Sarah The United States Air Force (USAF) requires a sustained and expanding pool of warranted Contracting Officers (COs) to meet growing operational and fiscal demands across its global enterprise. The authority to obligate funds and bind the government contractually—granted through the issuance of a warrant—requires successful completion of a multi-stage evaluation process culminating in a scenario-based oral board. This final interview assesses a candidate’s ability to interpret and apply acquisition policy under complex and ambiguous conditions. This paper proposes the design of an adaptive artificial intelligence (AI) training system—the Warrant Board Retrieval- Augmented Generation (RAG) Chatbot—to serve as a simulated board-preparation environment. This chatbot inverts the common ’user question, AI answer’ model, and instead has the chatbot ask the learner a series of critical thinking scenariobased questions. The prototype design adopts a model-agnostic LLM gateway capable of operation through either commercial APIs (e.g., OpenAI) or secure, government-hosted environments such as GenAI.mil, ensuring accessibility within unclassified Air Force networks. This research contributes to the emerging field of AI-assisted professional education by developing a transparent, auditable, and pedagogically grounded framework for formative learning in acquisition training.

AgentNexus: Accelerating AI Agent Development and Enhancing Interoperability with MCP

2026-02-17T00:00:00Z

AgentNexus: Accelerating AI Agent Development and Enhancing Interoperability with MCP Yae, Jung; Hamilton, Lei The DoD faces significant challenges in its pursuit of AI superiority, as disparate data and development platforms create redundant efforts and limit interoperability. Additionally, existing DoD systems are ill-equipped to handle the recent paradigm shift toward agentic AI, which requires modern standards and tools. To address these gaps, this paper introduces AgentNexus, an application designed to streamline the development, deployment, and servicing of AI agents. AgentNexus provides an application featuring an advanced agents processing backend, a scalable service layer, and an intuitive user interface. It provides pre-built toolkits, sophisticated RAG pipeline, and MCP for enhanced interoperability. The successful development of an Education Assistant agent validates the application’s capacity to support the rapid implementation of multi-agent workflows. By fostering a collaborative and standardized environment, AgentNexus mitigates critical barriers of interoperability and duplicated effort, accelerating the delivery of multi-agent AI to warfighters.

Intelligent C-17 Load Planning for Flight Optimization

2026-02-17T00:00:00Z

Intelligent C-17 Load Planning for Flight Optimization McAlister, Catherine; Jones, Mathew; McConville, Sean C-17 Globemaster III cargo capacity is significantly underutilized, with many sorties transporting only a few pallets despite the aircraft’s 170,900-pound payload capability. Historical flight data analysis reveals inefficient scheduling practices that increase operational costs, crew workload, and overall negatively effect mission capability. This paper details the development of an AI-powered optimization model to improve C-17 cargo utilization and reduce required flight operations. We analyzed historical C-17 transportation data and created both traditional optimization algorithms and predictive AI models to determine optimal flight scheduling for 3-week operational periods. The AI model achieved 97.9% accuracy in predicting optimal flight count requirements and 89.3% accuracy in predicting optimal flight assignment for specific cargo, representing a 23% reduction in total flights and a 15% increase in average cargo utilization. These results demonstrate that data-driven flight scheduling can significantly improve C-17 operational efficiency, reduce costs across the airlift community, and enabling additional time towards advanced training, contingency support, and critical warfighter operations, ultimately increasing the lethality and readiness of the Department of Defense.

Securing Intelligence: The Strategic Necessity of Air-Gapped AI Systems in the Age of Cloud-Based LLMs

2026-02-17T00:00:00Z

Securing Intelligence: The Strategic Necessity of Air-Gapped AI Systems in the Age of Cloud-Based LLMs Viggh, Herbert; Tsagaratos, Jennifer The increasing use of large language models (LLMs) in applications, from military strategy to customer service, raises concerns about data sovereignty, security, and privacy. Cloudbased API models, created by companies such as OpenAI, pose significant risks due to training data exposure and prompt injection attacks, which can compromise sensitive information and hidden biases that could influence reporting or executive decision-making processes. Real-world incidents, such as the leakage of Samsung’s proprietary source code through ChatGPT, highlight the dangers of relying on cloud providers with complete visibility into client queries. Furthermore, data localization laws and regulations, such as the General Data Protection Regulation (GDPR), underscore the risks associated with outsourcing intelligence and decision support systems to foreign entities. Airgapped AI solutions, which run on isolated networks disconnected from the outside world, offer a secure alternative for sensitive environments such as national defense, research laboratories, and critical infrastructure. By maintaining control over AI processes, organizations can ensure information safety, comply with regulations, and mitigate risks associated with cloud-based AI infrastructure, ultimately safeguarding their data integrity, privacy, and operational independence.

RAIMOND Requirements AI for Military Operational Needs Development

2026-02-17T00:00:00Z

RAIMOND Requirements AI for Military Operational Needs Development Garcia, Fabio; Steilberg, Jackson The Joint Capabilities Integration and Development System (JCIDS) was created as a means to overhaul military procurement processes. Ideally, the requirements development process is meant to take a total of 2-4 years from concept to manufacturing. However the actual length of concept development is much longer. As a result, technologies that are conceptualized through the analytical process often enter the acquisition too late to need for the warrior. To reduce the lengthy timeline in requirements development, we used Large Language Models (LLMs) to conduct the necessary research and synthesize documents that abide by strict JCIDS guidelines. Prompt engineering can achieve these results as a proof of concept. However, the output responses lack the content length and depth necessary to pass through the requirements validation process. Therefore, a combination of agentic workflows, prompt engineering, and sufficient context is needed to achieve the desired outcomes. This project utilizes a novel framework to derive Capabilities Based Assessments (CBAs) at an approximate 80 percent readiness level requiring the final steps of validation and verification by subject matter experts.

Machine Learning for the Enhancement of Adaptive Optics

2026-02-17T00:00:00Z

Machine Learning for the Enhancement of Adaptive Optics Hall, Robert; Chen, Justin Optical systems (telescopes, lasers, microscopes, etc.) have degraded performance over long distances due to scintillation caused by Earth’s atmosphere, where adaptive optics (AO) is often used to enhance its signal-to-noise (SNR) ratio or image quality. Astronomers have found success in laser-based adaptive optics where they survey the atmosphere with a laser and subtract its effects on the resultant image. Although effective in most cases, these systems can be extremely costly, are computationally intensive in real time, and fall short in some edge cases. We propose an autoencoder/ decoder and a generalized sequence to sequence model (LSTM) as a cost-effective method to off-load computational complexity from real time and enhance performance in edge cases. This study utilizes four simulated datasets of wavefront sensor frames for a variety of atmospheric conditions, done in collaboration with MIT Lincoln Laboratory [1]–found auto-encoding performance just shy of traditional methodology and found LSTM performance that predicts well the general shape on the WFS, but suffers from scaling issues.

From Hype to Reality: Real-World Lessons and Recommendations for AI in Military Applications

2026-02-17T00:00:00Z

From Hype to Reality: Real-World Lessons and Recommendations for AI in Military Applications Lynch, Joshua; Niss, Laura The current use cases, limitations, and future capacity of large language models (LLMs) as assistants to military personnel remain an open question. This paper presents a case study of an Airman’s interaction with and trust calibration of LLMs over three months, both as an everyday assistant and for development of ROMAD-AI, a tactical military application. Through intuitive, AI-generated software development, an approach relying on iterative code generation through natural language prompting of LLMs from a technical novice rather than human generated programming from a technical expert, the research reveals significant gaps between industry curated AI capability demonstrations and operational reality, requiring systematic trust calibration and realistic scope management. Outcomes are analyzed through operational and technical expertise perspectives to provide practical guidance for both military service members seeking effective AI integration and researchers developing military-focused AI systems.

Large Language Models and Defense Strategy: Escalation Risks and National Security Challenges

2026-02-17T00:00:00Z

Large Language Models and Defense Strategy: Escalation Risks and National Security Challenges Hou, Jonathan; Lax, Edwin This literature review examines the strategic vulnerabilities posed by Large Language Models (LLMs) in military and national security contexts. It synthesizes recent research on their propensity for escalatory reasoning, cultural misalignment, semantic manipulation, and dual-use ambiguity. Findings from conflict s imulations a nd c oalition p lanning m odels reveal how LLMs may default to aggressive or biased outputs under ambiguity. These tendencies threaten alliance cohesion, distort decision-making, and undermine trust in AI-enabled operations. The review concludes by advocating for safeguards such as culturally calibrated training, rigorous output verification, a nd the integration of human-AI intermediaries to prevent destabilizing outcomes.

Synchronization-Aware Diffusion Models for Intra-Family RF Signal Classification

2026-02-17T00:00:00Z

Synchronization-Aware Diffusion Models for Intra-Family RF Signal Classification Hayden, Hunter; Botero, Joey Classification of radio frequency (RF) signals in the presence of channel-induced synchronization errors remains a critical challenge in spectrum awareness systems. Traditional classification pipelines generally rely on fixed synchronization algorithms or assume aligned signals, which limits robustness under real world timing, phase, and frequency distortions. We introduce SyncDiff, a novel encoder-only diffusion model architecture that predicts synchronization parameters through iterative denoising steps prior to classification. By replacing conventional synchronization algorithms with a learned datadriven correction mechanism, our approach enables adaptive signal alignment based on current channel distortions in unsynchronized input data. SyncDiff employs a UNet based encoder to refine synchronization parameters across multiple inference steps, dynamically reducing channel-induced alignment errors while preserving the inherit modulation specific characteristics that allow these signals to be discriminable. Evaluations of the RadioML2018 RF standard benchmark data set [1] demonstrates improved classification accuracy across varying SNRs, modulation schemes and synchronization impairments. Our findings highlight the potential of diffusion-based synchronization learning to improve downstream RF classification without reliance on expert-engineered synchronization routines.

ML Prediction Models to Identify Novel Beyond Visual Range Tactics and Error Analysis for DARPA AIR Agents

2026-02-12T00:00:00Z

ML Prediction Models to Identify Novel Beyond Visual Range Tactics and Error Analysis for DARPA AIR Agents Li, William; Castor, Jeremy This paper investigates the utility of using machine learning models to predict the outcome of simulated 2 vs. 2 Tactical Intercept engagements flown by autonomous agents in support of the DARPA Artificial Intelligence Reinforcements (AIR) program. We investigated the performance of four models: Feed Forward Neural Network, Random Forest, Extreme Gradient Boost, and Long Short Term Memory (LSTM). We examined their ability to successfully predict the outcomes of simulated engagements, tactical errors, and the execution of novel game plans by autonomous agents. The models were trained on 53 features pertaining to the agents including distance between aircraft, altitude, speed, missile availability, and other eventbased features from simulated runs. The LSTM model had the best performance towards the beginning of a run and was able to predict the correct winner with 87.8% accuracy only one minute into a run while the XGBoost model achieved the best overall performance with a 91.7% classification accuracy and an R² of 0.712. The XGB model was also able to correctly predict the winner of 84.7% of the runs after only seven minutes into the simulated engagement. These results demonstrate the utility and need for further investigation into other ML models potential to identify unique attributes and predictive analysis of more complex multi-agent scenarios that include additional criteria such as varying rules of engagement, incorporating acceptable levels of risk as well as other requirements fighter pilots must take into account during offensive and defensive operations needed to gain air superiority and support the objectives of the Joint Forces Commander.

MIT hosts 11th Undergraduate Research Technology Conference

2025-10-30T00:00:00Z

MIT hosts 11th Undergraduate Research Technology Conference Beyah, Malakhi; Placides, Jojo From Oct. 10 to Oct. 12, the Stata Center was abuzz with bright minds and fresh faces as the Institute geared up for its 11th annual Undergraduate Research Technology Conference (URTC), where high school and undergraduate students from across the country came to present their latest research to experts and industry leaders.

Lincoln Laboratory and MIT Haystack Observatory partner to unveil hidden parts of the galaxy

2025-09-22T00:00:00Z

Lincoln Laboratory and MIT Haystack Observatory partner to unveil hidden parts of the galaxy Parde, Nathan They propose building a telescope made of thousands of tiny, identical satellites that will work together to reveal low-frequency radio waves in space.

DOE selects MIT to establish a Center for the Exascale Simulation of Coupled High-Enthalpy Fluid–Solid Interactions

2025-09-10T00:00:00Z

DOE selects MIT to establish a Center for the Exascale Simulation of Coupled High-Enthalpy Fluid–Solid Interactions Hadley, F The research center, sponsored by the DOE’s National Nuclear Security Administration, will advance the simulation of extreme environments, such as those in hypersonic flight and atmospheric reentry.

AI Accelerator Announces Award Winners

2025-07-31T00:00:00Z

AI Accelerator Announces Award Winners Accelerator, AI The Department of Air Force (DAF)-MIT AI Accelerator is a unique collaboration designed to advance the field of AI to improve DAF operations and address broader societal needs. In June 2025, the DAF-MIT Artificial Intelligence Accelerator named the recipients of AI Accelerator awards, recognizing scientific excellence, distinguished contributions, and other exceptional accomplishments. The awardees were nominated and selected from members of the AI Accelerator community, including individuals from the DAF, MIT campus, and MIT Lincoln Laboratory.

Responding to the Climate Impact of Generative AI

2025-09-30T00:00:00Z

Responding to the Climate Impact of Generative AI Zewe, Adam Explosive growth of AI data centers is expected to increase greenhouse gas emissions. Researchers are now seeking solutions to reduce these environmental harms.

Lincoln Lab Unveils the Most Powerful AI Supercomputer at any US University

2025-10-02T00:00:00Z

Lincoln Lab Unveils the Most Powerful AI Supercomputer at any US University Foy, Kylie Optimized for generative AI, TX-GAIN is driving innovation in biodefense, materials discovery, cybersecurity, and other areas of research and development.

Artificial Intelligence for Tactical Network Troubleshooting

2025-09-12T00:00:00Z

Artificial Intelligence for Tactical Network Troubleshooting Jaimes, Rafael; Mendez, Maximillian The tactical network is a key component of most United States Marine Corps missions. It is critical to expeditiously stand up a robust communications architecture for both voice and data transmissions across a variety of classification levels. However, when there are unforeseen or induced faults in network configurations, the establishment time can increase by hours if not days. The research described in this report sought to determine if a large language model (LLM), when provided the correct baseline network configurations, would be able to identify errors in active working network configurations and reduce network establishment time. A/B testing was conducted to see whether teams assisted by artificial intelligence (AI) or control teams with no AI assistance could establish the network faster. The LLM hosted by NIPRGPT decreased the establishment time by 50 percent (p <0.05) compared to warfighters unaided by AI. The results conclude that AI agents such as LLMs can be useful in providing commanders with a course of action to establish command, control, communications, and computers (C4) faster. The Department of the Air Force Artificial Intelligence Accelerator

A KNOWLEDGE GRAPH IS ALL YOU NEED

2025-09-10T00:00:00Z

A KNOWLEDGE GRAPH IS ALL YOU NEED Streilen, William; Brooks, Nicholas; Burill, Daniel; Smith, Corey The Department of the Air Force (DAF) faces unique challenges in adopting Large Language Models (LLMs). Commercially available models often lack the domain-specific knowledge necessary to support airmen, as this information is not inherently embedded. To maintain a competitive edge, the integration of LLMs to improve efficiency and decision making is a critical priority. This presentation explores two innovative methodologies designed to better integrate domain-specific knowledge into language models and improve the discovery of relevant information. The first is EntiGraph Continuous Pretraining, which leverages continuous training to embed specialized knowledge into language models. The second is the GFM-RAG Graph RAG Framework, a novel approach to knowledge retrieval and synthesis that enhances model performance by improving multi-hop retrieval and complex information connections. Through both quantitative and qualitative evaluations, we assess their impact on retrieval accuracy and response relevance. Our findings demonstrate the potential of these customized approaches to streamline information access, improve decision making, and better support the operational needs of the DAF.

LLM-Based Entity Extraction for Cyber Threat Reports

2025-09-10T00:00:00Z

LLM-Based Entity Extraction for Cyber Threat Reports Alperin, Kenneth; de Silva, Alexis As the cyber threat landscape and capabilities of advance persistent threats continue to expand, applying cutting-edge technology to the domain of cyber intelligence is necessary for the United States Space Force to keep pace in the Great Power Competition. Cyber intelligence analysts spend an estimated time of nearly 840 man-hours annually on the extraction and validation of relevant intelligence from cyber threat reports (CTRs). Named entity recognition (NER) is a natural language processing technique capable of automatically extracting and labeling all relevant information from a given text. Although not a novel idea, this paper aims to expand the current but limited research on the applications of NER to the domain of cyber intelligence. This study uses a new openly-licensed dataset, AnnoCTR, to finetune a cybersecurity-specific, transformers-based model, CYBERT. The performance of the model is compared to the models from the derived literature. Although the results showed an F1 score of 0.733 – a less optimal performance compared to previous models – there is still more work to explore to reduce the production time of intelligence analysis by half.

Democratizing Data: An Intelligent Querying System for Marine Corps Data

2025-09-10T00:00:00Z

Democratizing Data: An Intelligent Querying System for Marine Corps Data Johnson, Lane; Nam, Kevin This research presents the development and implementation of a text-to-Structured Query Language (SQL) system tailored for Marine Corps logistics, capitalizing upon the proven capabilities of Large Language Models (LLMs). By fine-tuning an open-source LLM on a curated Global Combat Support System - Marine Corps supply and maintenance dataset, we demonstrate how non-technical users can intuitively interact with Marine Corps data through natural language queries, enhancing data accessibility and operational decision-making. Our approach assumes a resource-constrained environment, demonstrating that fine-tuning and deploying the model on a single NVIDIA A100 graphics processing unit (GPU) is not only feasible, but also highlights the potential for local or edgebased artificial intelligence (AI) solutions. We further identify the critical importance of high-quality, representative datasets and propose a hybrid approach combining prompt engineering with fine-tuning to improve performance. Our findings culminate in concrete recommendations for the Marine Corps regarding data governance, AI integration, and workforce development.

Pixels to Places: Improving Zero-Shot Image Geolocalization using Prior Knowledge

2025-09-10T00:00:00Z

Pixels to Places: Improving Zero-Shot Image Geolocalization using Prior Knowledge Cha, Miriam; Borg, Trent The ability to predict the geographic origin of a photo is critical for open-source investigation applications. However, image geolocalization is highly challenging due to the vast diversity of images captured worldwide. While vision transformer-based approaches have demonstrated success— even outperforming grandmasters in geolocation games like GeoGuessr—their performance does not generalize well to unseen locations. Prior methods rely solely on visual cues, neglecting broader contextual knowledge that image analysts typically employ. To bridge this gap, our research integrates the contextual understanding of geographic regions that imagery analysts possess into the geolocalization model. Specifically, we develop a variant of StreetCLIP, which embeds CLIP within geolocalization tasks and facilitates the incorporation of user-supplied prior knowledge such as continental or national boundaries. Our results on the IM2GPS3K benchmark dataset demonstrate a 10.66% improvement in regional prediction (within 200 km) and a 15.27% improvement in country-level prediction (within 750 km) over baseline models. Our results suggest that humanprovided supervision can enhance image geolocalization accuracy, highlighting the potential of interactive systems where human expertise and AI work collaboratively to refine predictions. Index Terms—image geolocalization, CLIP, human-machine teaming, vision transformers

Improved Automatic Electronic Intelligence Collection System for Internal and External Forward Fusion and Collaborative Geolocation of Adversary Emitters

2025-09-10T00:00:00Z

Improved Automatic Electronic Intelligence Collection System for Internal and External Forward Fusion and Collaborative Geolocation of Adversary Emitters Botero, Joey; Benge, Arianne; Heisey, Curtis With the 2022 National Defense Strategy shifting focus from counterinsurgency operations to near-peer adversaries, airborne ISR platforms within the USAF and DoD must be improved for effectiveness in a near-peer conflict. They need to be able to operate quickly and effectively in contested environments with longer-range threats, act as a forward edge intelligence node for blue forces and provide DoD Research and Development efforts with cutting-edge data regarding new adversary signals and technology. To aid in tackling these challenges, this project introduces a Machine Learning (ML)-driven approach that revamps the Automatic Electronic Intelligence Collection System (ACS) on U.S. Airborne ISR platforms in four ways: First, by providing nodal analysis to the user in real time by automatically aggregating existing data across the aircraft to the user for decreased operator cognitive load. Second, increasing internal aircraft database information with external intelligence database information to increase confidence in targeting. Third, by providing automatic signal anomaly detection to the operator utilizing a support vector machines algorithm that cues operators to potential signals of interest based on previous activity and pattern of life prediction. Lastly, by providing better surface against airborne identification through utilization of cone angle to the system to help operators with faster threat warning and situational awareness of the environment. Findings include Support Vector Machines being the most effective tested binary classifier for predicting single signal anomaly detection at 84% AUC and a rule-based method of averages successfully classifying 1089 surface versus air ELINT samples with a success rate of 89% compared to other tested methods, such as Gaussian Mixture Models at 68% and KNearest Neighbor at 66%.

Artificial Intelligence for Derivative Security Classification: Applications to DoD

2025-09-10T00:00:00Z

Artificial Intelligence for Derivative Security Classification: Applications to DoD Gelbard, Andrew; Hamilton, Lei The accurate classification of government documents according to their sensitivity (e.g., UNCLASSIFIED, SECRET, TOP SECRET) is critical for national security, yet historically has relied on time-intensive manual review. The current manual classification process consumes millions of labor hours annually within the U.S. government, significantly diverting skilled personnel from essential analytical tasks. This research explores automating this security classification task using recently available declassified materials from the DISC dataset [1], addressing practical challenges such as noisy Optical Character Recognition (OCR) output, imbalanced data distributions, and potential leakage of explicit classification markers within document text. This dataset contains declassified government documents sourced from the Digital National Security Archive, providing authentic textual examples representative of actual classification scenarios. We evaluate both traditional machine learning approaches and advanced transformerbased language models to classify documents accurately across multiple sensitivity levels. Our results highlight that transformer-based models, particularly DeBERTa, effectively improve identification of the minority but critical TOP SECRET class, achieving recall over 70% and an overall balanced performance (macro F1 score of 0.75), while traditional methods exhibit similar overall accuracy but struggle with minority class recall. Despite promising findings, we caution that conclusions drawn here remain constrained by limited training data size and inherent uncertainties in human-labeled documents. We emphasize the need for larger, rigorously preprocessed datasets and suggest future research integrating authoritative classification guidelines directly into model training, potentially via retrieval-augmented methods. This work thus contributes a foundational, reproducible framework that demonstrates significant potential for machine-assisted security classification, guiding future research and practical applications in the information security domain.

The Area-of-Measurable-Performance (AOMP) Method Standard as a Foundational Archetype for the Cyclical Enhancement of the State of the Art Joint Simulation Environment (JSE) Technology

2025-09-10T00:00:00Z

The Area-of-Measurable-Performance (AOMP) Method Standard as a Foundational Archetype for the Cyclical Enhancement of the State of the Art Joint Simulation Environment (JSE) Technology Li, William; Johnson, Kevin; Picardo, Christopher; Ambion, Francis The Department of the Air Force (DAF) envisions the need to incorporate Artificial Intelligence and Machine Learning (AI/ML) models into novel systems it develops for the purpose of enhancing them to meet its primary goal of maintaining total air superiority [2]. There is currently a need for developing a standard process for the design of successful AI/ML models capable of enhancing the novel systems the DAF develops. In this white paper we introduce the Area of Measurable Performance (AOMP) Method Standard and apply it to the Joint Simulation Environment (JSE) Technology, a state of the art system of systems under test, to identify AOMPS and their modular requirements [3] and metrics that lead to the accurate characterization of modular AI/ML models through a process that offers a high degree of trust and reuse, resulting in a method standard that organically promotes the development of successful modular AI/ML models for use in the performance improvement of the JSE technology or other system of system(s) [4] under test.

Energy Innovation Through Research Computing at MGHPCC

2025-05-01T00:00:00Z

Energy Innovation Through Research Computing at MGHPCC Hill, Helen The Massachusetts Green High-Performance Computing Center (MGHPCC) is an innovation hub that supports research addressing critical energy and sustainability challenges. Research at the Center stands at the forefront of innovation, leveraging advanced computational techniques on state-of-the-art facilities to tackle some of society’s most pressing issues, particularly in energy and sustainability.

Artificial Intelligence Enhances Air Mobility Planning

2025-04-25T00:00:00Z

Artificial Intelligence Enhances Air Mobility Planning Foy, Kylie Lincoln Laboratory is transitioning tools to the 618th Air Operations Center to streamline global transport logistics.

Mission success for experiments in Mobility Guardian 23

2023-07-20T00:00:00Z

Mission success for experiments in Mobility Guardian 23 Sansano, Rachel Mobility Guardian 23 is giving U.S. and coalition forces an opportunity to experiment in the theater that matters.

New open-source tool helps to detangle the brain

2024-08-14T00:00:00Z

New open-source tool helps to detangle the brain McGovern, McGovern The software tool NeuroTrALE is designed to quickly and efficiently process large amounts of brain imaging data semi-automatically.

TBIRD technology could help image black holes’ photon rings

2024-11-19T00:00:00Z

TBIRD technology could help image black holes’ photon rings Tantillo, Ariana The Lincoln Laboratory-developed laser communications payload operates at the data rates required to image these never-before-seen thin halos of light.

DAF AI Accelerator’s Fundamental Research Provides Breakthrough for Aircrew Readiness

2024-04-30T00:00:00Z

DAF AI Accelerator’s Fundamental Research Provides Breakthrough for Aircrew Readiness The Department of the Air Force Artificial Intelligence Accelerator Program developed a new tool, the Puckboard Intelligent Recommendation Engine, which has the potential to improve mission readiness through smarter, more efficient aircrew mission and training scheduling.

New AI tool generates realistic satellite images of future flooding

2024-11-25T00:00:00Z

New AI tool generates realistic satellite images of future flooding Chu, Jennifer A generative AI model visualizes how floods in Texas would look like in satellite imagery.

Q&A: The climate impact of generative AI

2025-01-13T00:00:00Z

Q&A: The climate impact of generative AI McGovern, Anne As the use of generative AI continues to grow, Lincoln Laboratory's Vijay Gadepally describes what researchers and consumers can do to help mitigate its environmental impact. Vijay Gadepally, a senior staff member in the Lincoln Laboratory Supercomputing Center, discusses steps the research community can take to help mitigate the environmental impact of generative AI.

Fifteen Lincoln Laboratory Technologies Receive 2024 R&D 100 Awards

2024-09-24T00:00:00Z

Fifteen Lincoln Laboratory Technologies Receive 2024 R&D 100 Awards Foy, Kylie The innovations map the ocean floor and the brain, prevent heat stroke and cognitive injury, expand AI processing and quantum system capabilities, and introduce new fabrication approaches.

How AI is improving simulations with smarter sampling techniques

2024-10-02T00:00:00Z

How AI is improving simulations with smarter sampling techniques Gordon, Rachel MIT CSAIL researchers created an AI-powered method for low-discrepancy sampling, which uniformly distributes data points to boost simulation accuracy.

New Tools for Navigating the Highways of Internet Traffic

2024-10-28T00:00:00Z

New Tools for Navigating the Highways of Internet Traffic Ramdhony, Vaneshi We live in a world where the internet connects everything from global economies to smart homes. Just like how people travel from one place to another by different modes of transport and across different streets, data also travels from one point to another through different containers and takes different routes. Engineers use network analysis to learn more about all this digital traffic.

“Better Networks” Project is Working to Improve Cybersecurity

2024-11-25T00:00:00Z

“Better Networks” Project is Working to Improve Cybersecurity Comeau, Kailen In collaboration with the MIT Sociotechnical Systems Research Center and the Air Force, the Lincoln Laboratory Supercomputing Center is working to develop better sensors using Laboratory developed Dynamic Distributed Dimensional Data Model (D4M) technology and artificial intelligence algorithms to support defensive cyber operations.

How MIT’s Rad Lab rescued D-Day

2024-10-22T00:00:00Z

How MIT’s Rad Lab rescued D-Day Fine, Norman After two British physicists invented a revolutionary gadget, MIT researchers used it to develop the radar devices that helped defeat the Nazis.

The Lincoln Scholars and Military Fellows Programs Foster Collaboration and Research to Prepare for the Future

2024-10-31T00:00:00Z

The Lincoln Scholars and Military Fellows Programs Foster Collaboration and Research to Prepare for the Future Ornitz, Rachel

MIT Is Developing an AI Co-Pilot for Aircraft Called Air-Guardian

2023-10-11T00:00:00Z

MIT Is Developing an AI Co-Pilot for Aircraft Called Air-Guardian Dela Cruz |, Jace This new tech serves as a proactive co-pilot, enhancing safety during critical moments of flight.

An AI Dataset Carves New Paths to Tornado Detection

2024-04-16T00:00:00Z

An AI Dataset Carves New Paths to Tornado Detection Foy, Kylie Tornadoes are violent mysteries. A public artificial intelligence dataset could help models reveal when and why they form, improving forecasters' ability to issue warnings.

New Open-source Tool Helps to Detangle the Brain

2024-07-22T00:00:00Z

New Open-source Tool Helps to Detangle the Brain McGovern, Anne

Study: Transparency is Often Lacking in Datasets Used to Train Large Language Models

2024-08-30T00:00:00Z

Study: Transparency is Often Lacking in Datasets Used to Train Large Language Models Zewe, Adam Researchers developed an easy-to-use tool that enables an AI practitioner to find data that suits the purpose of their model, which could improve accuracy and reduce bias.

Exploring the Mysterious Alphabet of Sperm Whales

2024-10-31T00:00:00Z

Exploring the Mysterious Alphabet of Sperm Whales Gordon, Rachel MIT CSAIL and Project CETI researchers reveal complex communication patterns in sperm whales, deepening our understanding of animal language systems.

William Green named director of MIT Energy Initiative

2024-05-06T00:00:00Z

William Green named director of MIT Energy Initiative Liberty, Janine In his new role, the professor of chemical engineering plans to speed up the consensus process among academics, business leaders, and policymakers for a successful energy transition.

Accelerating Fielding of Artificial Intelligence Programs with the DAF-MIT AI Accelerator

2024-04-19T00:00:00Z

Accelerating Fielding of Artificial Intelligence Programs with the DAF-MIT AI Accelerator Foy, Kylie The Department of the Air Force-Massachusetts Institute of Technology Artificial Intelligence Accelerator (DAF-MIT AI Accelerator) is a collaboration between multiple organizations, and aims to leverage the capabilities of academia, government, and industry to make progress in the field of artificial intelligence. Staff from the Laboratory work closely with scientists in government and across MIT campus on the Automation in Space Domain Awareness project—the first and currently only—U.S. Space Force AI accelerator project.

Using Deep Learning to Image the Earth’s Planetary Boundary Layer

2024-03-15T00:00:00Z

Using Deep Learning to Image the Earth’s Planetary Boundary Layer Wahl, Haley Lincoln Laboratory researchers are using AI to get a better picture of the atmospheric layer closest to Earth's surface. Their techniques could improve weather and drought prediction.

The Cloud Outgrows Linux, and Sparks a New Operating System

2024-03-12T00:00:00Z

The Cloud Outgrows Linux, and Sparks a New Operating System Prickett Morgan, Timothy Ultimately, every problem in the constantly evolving IT software stack becomes a database problem, which is why there are 418 different databases and datastores in the DB Engines rankings and there are really only a handful of commercially viable operating systems. But what if the operating system is the problem?

Lincoln Laboratory kicks off collaboration with Air University, Air Command and Staff College

2024-03-06T00:00:00Z

Lincoln Laboratory kicks off collaboration with Air University, Air Command and Staff College Wahl, Haley Students in the Joint All-Domain Strategist Graduate Program are embedded within project teams to discover the art of the possible when faced with technical challenges.

RAAINS workshop considers the future of artificial intelligence

2024-02-09T00:00:00Z

RAAINS workshop considers the future of artificial intelligence Lincoln Laboratory hosts its fifth-annual workshop focused on AI for national security.

Researchers release open-source space debris model

2024-02-09T00:00:00Z

Researchers release open-source space debris model The MIT Orbital Capacity Assessment Tool lets users model the long-term future space environment.

Sequence Evaluation in Real Time

2023-12-19T00:00:00Z

Sequence Evaluation in Real Time The ability to identify and characterize the function of an unknown pathogen, such as a virus, bacteria, or toxin, from its biological sequence is critical to quickly determine the potential impact against human health. SEQer aims to build an end-to-end computational testbed that can stream biological sequences in real-time and determine their function to identify possible biothreats.

Poisoning Cyberattacks to Design of Artificial Intelligence

2023-12-18T00:00:00Z

Poisoning Cyberattacks to Design of Artificial Intelligence The MIT Lincoln Laboratory project Poisoning Cyberattacks to Design of Artificial Intelligence (PoCyDAIn) aims to develop a framework for assessing the impact of poisoning attacks on cyber-ML systems.

Multimodal Vision for 3D Scene Interpretation

2023-12-18T00:00:00Z

Multimodal Vision for 3D Scene Interpretation The MIT Lincoln Laboratory project Multimodal Vision for 3D Scene Interpretation (MMV3D) translates multiple types of 2D imagery to create a 3D image that more accurately captures complex multi-surface geometry.

Award-Winning Laboratory Technology Aims to Prevent Aircraft Collisions

2023-12-18T00:00:00Z

Award-Winning Laboratory Technology Aims to Prevent Aircraft Collisions A next-generation collision avoidance system will help pilots and unmanned aircraft safely navigate the airspace.

Using Machine Learning to Trace Genetically Engineered DNA

2023-12-18T00:00:00Z

Using Machine Learning to Trace Genetically Engineered DNA A Laboratory team is using machine learning to trace the origin of DNA modifications. While previous studies focused on using plasmids—extra-chromosomal pieces of DNA—this team focused on trying to pinpoint exact computational tools used for editing the genome. The results of their work show that it may be possible to trace the origin of modifications back to a specific program, which may help identify the culprit in any attack involving genetic modification.

AI Accelerates Problem-solving in Complex Scenarios

2023-12-05T00:00:00Z

AI Accelerates Problem-solving in Complex Scenarios Zewe, Adam Researchers from MIT and ETH Zurich have developed a new, data-driven machine-learning technique that could be applied to many complex logistical challenges, such as package routing, vaccine distribution, and power grid management.

Combining Neural Networks and Histogram Layers for Underwater Target Classification

2023-12-18T00:00:00Z

Combining Neural Networks and Histogram Layers for Underwater Target Classification To detect targets, sea vessels largely rely on passive sonar, which records sounds with an underwater microphone. However, techniques for processing and analyzing passive sonar data often struggle to disentangle the complex patterns in target recordings. To better capture statistical features within passive sonar data, a team from Lincoln Laboratory and the Advanced Vision and Learning Lab at Texas A&M University are adding local histogram layers into neural network architectures. This project employs two types of neural networks for automated feature learning that together can capture local relationships within audio signals while incorporating signal time dependencies.

The Soviet Atomic Threat, Oppenheimer, and the Need for National Air Defense

2023-12-14T00:00:00Z

The Soviet Atomic Threat, Oppenheimer, and the Need for National Air Defense Widespread concern about the potential of a Soviet Union nuclear bomber threat prompted the Department of Defense (DoD) to initiate studies to evaluate the nation’s state of air defense against nuclear attack. This ultimately led to the creation of Project Lincoln and the establishment of Lincoln Laboratory. In the summer of 1952, a group of scientists, engineers, and military personnel met at Lincoln Laboratory to consider ways to improve the air defense of North America. Physicist J. Robert Oppenheimer visited the Laboratory to participate in this 1952 Summer Study, as did a number of other distinguished scientists. Oppenheimer became very concerned about the threat of air invasion after the summer study, and was a strong proponent for the programs established at Lincoln Laboratory.

New tools are available to help reduce the energy that AI models devour

2023-10-05T00:00:00Z

New tools are available to help reduce the energy that AI models devour Foy, Kylie The MIT Lincoln Laboratory Supercomputing Center (LLSC) is developing techniques to help data centers reel in energy use. Their techniques range from simple but effective changes, like power-capping hardware, to adopting novel tools that can stop AI training early on. One of their techniques can reduce the energy of training AI models by 80 percent. Their work is mobilizing green-computing research and promoting a culture of transparency.

Ancient Amazonians Intentionally Created Fertile “Dark Earth”

2023-09-20T00:00:00Z

Ancient Amazonians Intentionally Created Fertile “Dark Earth” Chu, Jennifer The Amazon river basin is known for its immense and lush tropical forests, so one might assume that the Amazon’s land is equally rich. In fact, the soils underlying the forested vegetation, particularly in the hilly uplands, are surprisingly infertile. Much of the Amazon’s soil is acidic and low in nutrients, making it notoriously difficult to farm. But over the years, archaeologists have dug up mysteriously black and fertile patches of ancient soils in hundreds of sites across the Amazon. This “dark earth” has been found in and around human settlements dating back hundreds to thousands of years. And it has been a matter of some debate as to whether the super-rich soil was purposefully created or a coincidental byproduct of these ancient cultures. Now, a study led by researchers at MIT, the University of Florida, and in Brazil aims to settle the debate over dark earth’s origins.

What Would You Do With a 16.8 Million Core Graph Processing Beast?

2023-09-01T00:00:00Z

What Would You Do With a 16.8 Million Core Graph Processing Beast? Morgan, Timothy Prickett The US Defense Advanced Research Projects Agency has been looking into creating a massively parallel graph processor and interconnect since establishing the Hierarchical Identify Verify Exploit (HIVE) project back in 2017. Intel was chosen to make the HIVE processor and Lincoln Laboratory at MIT and Amazon Web Services were chosen to create and host a trillion-edge graph dataset for a system based on such processors to chew on.

Simple Data Architecture Best Practices for AI Readiness

2023-11-09T00:00:00Z

Simple Data Architecture Best Practices for AI Readiness Gadepally, Vijay; Kepner, Jeremy AI1 requires data. A core requirement for AI techniques to be successful is high quality data. Hence, preparing systems to be “AI Ready” involves collecting raw data and parsing it. There are simple techniques that can be applied during initial parsing of raw data that can dramatically reduce the effort of applying AI. This document provides a short list of a few best practices for preparing the data.

Developing Artificial Intelligence for Noncooperative Space Operations Using Kerbal Space Program

2023-10-05T00:00:00Z

Developing Artificial Intelligence for Noncooperative Space Operations Using Kerbal Space Program With nearly 9,000 active and inactive satellites in orbit, space has become more crowded and competitive than ever before. Solving nascent problems in this domain — such as avoiding collision with debris or servicing malfunctioning satellites — requires more than just knowledge of orbital mechanics and spacecraft control. In the popular video game Kerbal Space Program (KSP), a team at the Laboratory saw a sophisticated multi-physics simulator capable of modeling all aspects of the aerospace domain. The team used the game engine to develop an open-source library and challenge suite, Kerbal Space Program Differential Games (KSPDG), designed to spur development of AI for a wide range of problems within the orbital domain.

Newly discovered planet has longest orbit yet detected by the TESS mission

2023-08-30T00:00:00Z

Newly discovered planet has longest orbit yet detected by the TESS mission Chu, Jennifer More than 80 percent of confirmed exoplanets have orbits shorter than 50 days. Astronomers are starting to get a general picture of these planets’ formation, evolution, and composition. But the picture is much fuzzier for planets with longer orbital periods. Far-out worlds, with months- to years-long orbits, are more difficult to detect, and their properties have therefore been trickier to discern. Now, the list of long-period planets has gained two entries. Astronomers at MIT, the University of New Mexico, and elsewhere have discovered a rare system containing two long-period planets orbiting TOI-4600, a nearby star that is 815 light years from Earth. The discovery was made using data from NASA’s Transiting Exoplanet Survey Satellite, or TESS — an MIT-led mission that monitors the nearest stars for signs of exoplanets. The new, farther planet has the longest period that TESS has detected to date.

ACAS X Provides Next-Generation Collision Avoidance

2023-09-28T00:00:00Z

ACAS X Provides Next-Generation Collision Avoidance ACAS X is a family of collision avoidance systems that use machine learning to optimize traffic alerts, resulting in dramatically reduced nuisance alerting and improved safety.

Einblick Launches Deep AI Integration, Empowering Users to Go from Prompts to Multimodal Data Workflows

2023-07-24T00:00:00Z

Einblick Launches Deep AI Integration, Empowering Users to Go from Prompts to Multimodal Data Workflows Einblick recently released Einblick Prompt, bringing the power of OpenAI and ChatGPT straight into their canvas-based data notebooks. Prompt is an AI agent that reasons and solves users’ natural language prompts, allowing them to generate and debug code, create charts, and build machine learning models in seconds.

Roadmap for Uncertainty: Implementing Modernization and Operationalizing Change

2023-01-27T00:00:00Z

Roadmap for Uncertainty: Implementing Modernization and Operationalizing Change Sneider, Ethan The Phantom Fellowship Program at the Department of the Air Force-Massachusetts Institute of Technology Artificial Intelligence Accelerator (DAF-MIT AIA) is a great case study for how to be successful in new environments. The resulting mental model titled “The Roadmap for Uncertainty,” as derived from the Phantom Program, shows how success is a simple matter of having the right tools readily available.

First Commercial Company Leverages DAF-MIT Maneuver ID Challenge to Advance AI

2023-01-18T00:00:00Z

First Commercial Company Leverages DAF-MIT Maneuver ID Challenge to Advance AI The Department of the Air Force-MIT Artificial Intelligence Accelerator (AIA) announced that Crowdbotics is the first company to successfully leverage the Maneuver Identification Challenge to advance the field of AI. The DAF-MIT AIA Maneuver ID Challenge is an open challenge designed to enable AI coaching and automatic maneuver grading in pilot training.

Security leaders learn AI fundamentals through MIT professional program

2022-05-20T00:00:00Z

Security leaders learn AI fundamentals through MIT professional program Chase, Brittany The AI Accelerator partnered with MIT Lincoln Laboratory, DOD’s Chief Digital and Artificial Intelligence Office, and MIT’s Computer Science and Artificial Intelligence Laboratory to offer the AI for National Security Leaders course, teaching defense officials the basics of modern AI. The three-day training opportunity was designed to inform DOD senior leaders about AI’s national security implications and ways to apply it in their domains of responsibility.

Phantom Fellowship Program Culminates in BAH Optimization Project

2023-01-17T00:00:00Z

Phantom Fellowship Program Culminates in BAH Optimization Project The Phantom Fellowship is a rigorous program for Airmen and Guardians interested in gaining exposure to AI/ML technologies. Part of the Phantom Fellowship Program mission is to develop advocates of AI within the DoD to help the rest of the services understand the capabilities and limitations of AI. Participants of a recent capstone project in the program with the Department of the Air Force-Massachusetts Institute of Technology Artificial Intelligence Accelerator examined possible solutions using AI and Machine Learning models to support and augment the process of calculating Basic Allowance for Housing.

MagNav Project Successfully Demonstrates Real-Time Magnetic Navigation

2023-05-26T00:00:00Z

MagNav Project Successfully Demonstrates Real-Time Magnetic Navigation In a groundbreaking achievement, the Department of the Air Force-Massachusetts Institute of Technology Artificial Intelligence Accelerator (AIA) MagNav project performed real-time magnetic navigation (MagNav) on the C-17A Globemaster III in flight, becoming the first organization to successfully demonstrate this cutting-edge technology in real-time on a Department of Defense aircraft. The successful demonstration of MagNav marks a significant milestone in advancing navigation capabilities for the U.S. Air Force.

A Future of Unmanned Aerial Vehicles

2023-09-28T00:00:00Z

A Future of Unmanned Aerial Vehicles Though UAVs have continued to be produced, regulations have not been modified since they were first outlined in 2006. Now that these vehicles are being used for tasks such as package delivery, medical supply delivery, and infrastructure inspection, the need for regulations regarding them has become more pressing. Ngaire Underhill, Associate Staff member at MIT Lincoln Laboratory, Surveillance Systems group, has spent much time thinking about these issues in collaboration with the FAA.

MIT and Stanford researchers develop operating system with one major promise: Resisting ransomware

2023-04-21T00:00:00Z

MIT and Stanford researchers develop operating system with one major promise: Resisting ransomware Vazquez, Christian Computer science researchers at MIT and Stanford are developing an operating system with built-in cybersecurity defenses.

Laboratory Study Compares Machine Learning Accelerator Technologies

2023-09-28T00:00:00Z

Laboratory Study Compares Machine Learning Accelerator Technologies Dr. Albert Reuther, Senior Staff, Lincoln Laboratory Supercomputing Center (LLSC), is leading a Laboratory study that investigates and compares machine learning accelerators to better understand each accelerator’s strengths and weaknesses. With a better understanding of how accelerators lead to increased performance, the LLSC is able to advise the Department of Defense to make more informed decisions about which machine accelerators to utilize in their computer systems.

Phiala Shanahan is seeking fundamental answers about our physical world

2023-02-28T00:00:00Z

Phiala Shanahan is seeking fundamental answers about our physical world Chu, Jennifer With supercomputers and machine learning, MIT physicist Phiala Shanahan aims to illuminate the structure of everyday particles and uncover signs of dark matter.

National Program Provides Real Cybersecurity Experience

2023-02-03T00:00:00Z

National Program Provides Real Cybersecurity Experience Smith, Kaye A team of University of Arizona graduate students was matched in spring 2022 with technical directors Jeremy Kepner and Hayden Jananthan from MIT Lincoln Laboratory to better identify the cyber characteristics of large-scale network data in order to help organizations defend cyberattacks.

Top HPC Players Creating New Security Architecture Amid Neglect

2023-01-20T00:00:00Z

Top HPC Players Creating New Security Architecture Amid Neglect Shah, Agam The concerns over security of high-performance computers are now being taken seriously in both the public and private sector, who are jointly defining a security architecture as part of a working group called High-Performance Computing Security, which is managed by National Institute of Standards and Technology and the National Science Foundation.

Program teaches US Air Force personnel the fundamentals of AI

2023-01-11T00:00:00Z

Program teaches US Air Force personnel the fundamentals of AI Zewe, Adam MIT researchers launched a pilot program to develop and study an AI education program that could provide essential skills to Air Force personnel with varied backgrounds and job requirements.

Parallel Thermal Analysis on LLSC

2023-04-06T00:00:00Z

Parallel Thermal Analysis on LLSC The Lincoln Laboratory Supercomputing Center is exploring more designs and solving larger space thermal models using the supercomputing workflow developed by the Engineering Research Technical Investment project. The project integrated three key components for space systems design: a powerful thermal solver, Lincoln Laboratory’s Integrated Modeling and Analysis Software (LLIMAS), and the computing resources on the LLSC platform.

Two Laboratory Teams Receive 2020 AI Accelerator Awards

2022-11-10T00:00:00Z

Two Laboratory Teams Receive 2020 AI Accelerator Awards Two teams involving Laboratory staff were recognized with 2020 awards from the Department of the Air Force (DAF)−MIT Artificial Intelligence (AI) Accelerator. Recipients were selected from more than 150 airmen, researchers, faculty, and students from the DAF, the Laboratory, and MIT who are part of a collaborative effort to accelerate fundamental advances in AI. The AIA Director's Award, which highlights "excellence and impact with a focus on collaboration across the AIA and with stakeholders," recognized the Earth Intelligence Engine (EIE) project team for "cross-organizational collaboration, curation of novel datasets, visualization of forecasts, and delivery of an innovative challenge problem." The Earth Intelligence Engine (EIE) project targets three research areas — the earth intelligence platform, earth intelligence enhancement, and earth visual models — to build weather and climate resiliency for the U.S. Air Force. Resiliency against hurricanes, wildfires, flooding, sea level rise, and other extreme-weather and climate related threats is critical to protecting USAF resources such as military bases and ensuring mission readiness. The AIA Challenge Award recognized the Puckboard project team for designing and implementing two community challenge problems focused on aircrew scheduling. Puckboard is a web-based software application for scheduling pilots and loadmasters—personnel responsible for loading and unloading cargo and passengers—to mission and training flights.

Generating New Molecules with Graph Grammar

2022-04-01T00:00:00Z

Generating New Molecules with Graph Grammar Hinkel, Lauren An efficient machine-learning method uses chemical knowledge to create a learnable grammar with production rules to build synthesizable monomers and polymers.

Green AI Architecture Experimentation

2022-11-10T00:00:00Z

Green AI Architecture Experimentation The Green AI Architecture Experimentation (GAIA-X) project is developing technologies to promote the concept of green computing by establishing foundational tools and proposing power-reduction strategies. The immediate goals center on developing intelligent tools and strategies to measure and track power consumption in the data center, and demonstrating the effectiveness of various power reduction strategies on AI training and experimentation.

Laboratory Technology Helps Manage Flight Delays Through Severe Weather

2022-11-10T00:00:00Z

Laboratory Technology Helps Manage Flight Delays Through Severe Weather Historically, air traffic controllers (ATCs) have relied on past experience and looked at weather forecasts themselves to understand and prepare in advance for weather that could impact the flow of traffic. A MIT Lincoln Laboratory-developed technology called the Traffic Flow Impact (TFI) tool aims to help ATCs make detailed plans in advance using machine learning. The tool, which began development in 2013 and recently won an R&D 100 Award, utilizes multiple weather forecast models from the National Oceanic and Atmospheric Administration and machine learning to analyze the different forecast models. It also considers historical data of how traffic has been interrupted by weather in order to provide a breakdown of how ATCs may want to adjust the flow of traffic.

Taking a magnifying glass to data center operations

2022-08-24T00:00:00Z

Taking a magnifying glass to data center operations Foy, Kylie To gain insight into whether the MIT Lincoln Laboratory Supercomputing Center (LLSC) TX-GAIA supercomputer is being used as effectively as it can, the LLSC has been collecting detailed data on system usage over the past year. More than a million user jobs later, the team has released the dataset open source to the computing community. The goal is to empower computer scientists and data center operators to better understand avenues for data center optimization.

Intelligent Tornado Prediction Engine

2022-11-09T00:00:00Z

Intelligent Tornado Prediction Engine The most common type of tornado in the Southeast region of the United States, known as a quasi-linear convective system tornado, is historically difficult to warn for, with lead times hovering under 7 minutes and a false alarm rate of over 75 percent. The Intelligent Tornado Prediction Engine (ITORPE) combines meteorological and machine learning expertise from MIT Lincoln Laboratory researches and the Lincoln Laboratory Supercomputing Center to perform extremely large-scale data fusion spanning several years’ worth of radar, satellite, model, and in situ observation platforms, to provide enhanced situational awareness to forecasters using a graphical interface to focus forecasters’ attention on the storms of highest importance.

GraphBLAS and GraphChallenge Advance Network Frontiers

2022-11-09T00:00:00Z

GraphBLAS and GraphChallenge Advance Network Frontiers Kepner, Jeremy; Bader, David A.; Davis, Tim; Pearce, Roger; Wolf, Michael M. The challenges associated with graph algorithm scaling led multiple scientists to identify the need for an abstraction layer that would allow algorithm specialists to write high-performance, matrix-based graph algorithms that hardware specialists could then design to without having to manage the complexities of every type of graph algorithm. With this philosophy in mind, a number of researchers (including two Turing Award winners) came together and proposed the idea that “the state of the art in constructing a large collection of graph algorithms in terms of linear algebraic operations is mature enough to support the emergence of a standard set of primitive building blocks”

Laboratory Workshops Highlight Use of Artificial Intelligence in National Security

2022-04-05T00:00:00Z

Laboratory Workshops Highlight Use of Artificial Intelligence in National Security Teams at MIT Lincoln Laboratory are using AI to address national security challenges in collaboration with other organizations. To facilitate this research and promote collaboration with other organizations, the Laboratory virtually hosted the third annual Recent Advances in Artificial Intelligence for National Security (RAAINS) and Human-Machine Collaboration for National Security (HMC) workshops. The theme of this year’s HMC workshop was “Human-Centered AI.” The Department of Defense (DoD) and other national security organizations have recently highlighted this as a crucial focus for AI research and development, and also have called for a focus on AI ethics and reliability as AI increasingly becomes an equal partner in important national security work.

Technology Office Announces Winners of the FoolMe Hackathon

2022-04-05T00:00:00Z

Technology Office Announces Winners of the FoolMe Hackathon The MIT Lincoln Laboratory Technology Office wrapped up the FoolMe Challenge, a hackathon that was part of the Laboratory’s ongoing effort to observe trends in the manipulation of information, anticipate the broader implications to national security, and develop mitigation strategies. Teams worked to develop new methods of detecting manipulated images in six datasets that had been modified using different data poisoning techniques. The teams were judged based on the correct identification of manipulated images and the novelty of their problem-solving approach.

Lincoln Laboratory Supercomputing Center 5th Anniversary

2022-01-04T00:00:00Z

Lincoln Laboratory Supercomputing Center 5th Anniversary The Lincoln Laboratory Supercomputing Center (LLSC) celebrated a significant anniversary in 2021, marking five years of the center’s mission to enhance the computing power available to the Laboratory, MIT, and other researchers. This 5th anniversary compendium of articles written by the dedicated staff at the Lincoln Laboratory Bulletin, MIT News, and other writers is a testament to the continuing impact and potential of supercomputing to advance science and engineering.

MIT Researchers Turn to Unity 3D Game Engine for Supercomputer Diagnostics

2021-10-29T00:00:00Z

MIT Researchers Turn to Unity 3D Game Engine for Supercomputer Diagnostics Kostovic, Aleksandar To speed up the process of monitoring, diagnosing, and fixing problems with multi-billion-dollar supercomputers researchers from the Laboratory and MIT have developed a new technology to visualize node monitoring, offering real-time system reporting in the Unity 3D game engine found in many video games.

Air Force Pilots Get an AI-Assist With Scheduling Aircrews

2021-07-08T00:00:00Z

Air Force Pilots Get an AI-Assist With Scheduling Aircrews Foy, Kylie Scheduling C-17 aircraft crews is complicated. It’s a pain point for Airmen of 52 squadrons who operate C-17s, the military cargo aircraft that transport troops and supplies globally. This year, the Air Force marked four million flight hours for its C-17 fleet, which comprises 275 U.S. and allied aircraft. Each flight requires scheduling a crew of six on average. A team spanning MIT Lincoln Laboratory, the Department of the Air Force and the MIT Department of Aeronautics and Astronautics collaborated with their Air Force sponsor organization to develop an AI–enabled plugin for the existing C-17 scheduling tool that automates C-17 aircrew scheduling and optimizes crew resources.

MIT Lincoln Laboratory Takes the Mystery Out of Supercomputing

2017-01-18T00:00:00Z

MIT Lincoln Laboratory Takes the Mystery Out of Supercomputing In this special guest feature, Dr. Jeremy Kepner from MIT Lincoln Laboratory describes the lab’s approach to developing algorithms that will keep their users productive as new processing technologies evolve.

MIT Lincoln Lab Offers Advice for Delivering On-Demand HPC

2019-03-11T00:00:00Z

MIT Lincoln Lab Offers Advice for Delivering On-Demand HPC Russell, John Looking for advice on how to deliver HPC to a diverse science user community? MIT’s Lincoln Laboratory has just posted a new paper – Lessons Learned from a Decade of Providing Interactive, On-Demand High Performance Computing to Scientists and Engineers – intended to fill the bill.

Capitalizing on Machine Learning - from Life Sciences to Financial Services

2016-12-26T00:00:00Z

Capitalizing on Machine Learning - from Life Sciences to Financial Services The promise of machine learning has a science fiction flavor to it: computer programs that learn from their experiences and get better and better at what they do. So is machine learning fact or fiction? The global marketplace answers this question emphatically: Machine learning is not just real; it is a booming field of technology that is being applied in countless artificial intelligence (AI) applications. In the life sciences arena, researchers are leveraging machine learning in their work to drive groundbreaking discoveries that may help improve the health and wellbeing of people.

Supercharging Big Data Research with New England’s Fastest Supercomputer

2017-01-11T00:00:00Z

Supercharging Big Data Research with New England’s Fastest Supercomputer ISTC for Big Data caught up with ISTC for Big Data Principal Investigator and Lincoln Laboratory fellow Dr. Jeremy Kepner, who heads the LLSC, to learn more about Lincoln Laboratory's new supercomputer, how it’s helping ISTC research, and his work for the ISTC.

Hydra-zen Framework Makes Scientific Computing Easier for Researchers

2021-09-01T00:00:00Z

Hydra-zen Framework Makes Scientific Computing Easier for Researchers Hydra-zen is a framework that allows researchers to better document and implement their changes to variables and settings used to perform complex experiments and save them alongside the resulting calculations, ensuring that the experiment can be duplicated. Hydra-zen aims to simplify and automate the scientific computing process.

2020 IEEE Boston Section Distinguished Service Award Recipient

2021-08-23T00:00:00Z

2020 IEEE Boston Section Distinguished Service Award Recipient The 2020 recipient of the IEEE Boston Section’s “Distinguished Service Award” (DSA) is Dr. Albert Reuther.

Award-Winning Tools Enable Agencies to Efficiently and Easily Analyze Surveillance Footage

2021-08-09T00:00:00Z

Award-Winning Tools Enable Agencies to Efficiently and Easily Analyze Surveillance Footage Closed-circuit television and other surveillance systems are commonplace in busy, high-traffic areas. However, when investigators need to use these systems to search through recorded footage, these systems’ clunky interfaces and tools can create hours of work just in organizing footage. The Forensic Video Exploitation and Analysis (FOVEA) suite of tools, developed at the Laboratory in response to the Boston Marathon bombing in 2013, is an easy to set up and use solution to this problem.

MIT joins White House supercomputing effort to speed up search for Covid-19 solutions

2020-03-23T00:00:00Z

MIT joins White House supercomputing effort to speed up search for Covid-19 solutions Chu, Jennifer MIT joins a consortium of supercomputing facilities to help speed the search for COVID-19 solutions.

Dr. Jeremy Kepner Named Vice Chair of the Society for Industrial and Applied Mathematics (SIAM) Data MIning and Analytics Activity Group

2021-08-02T00:00:00Z

Dr. Jeremy Kepner Named Vice Chair of the Society for Industrial and Applied Mathematics (SIAM) Data MIning and Analytics Activity Group Dr. Jeremy Kepner has been elected vice chair of the Society for Industrial and Applied Mathematics (SIAM) Activity Group on Data Mining and Analytics (SIAG/DMA) for the 2014-2015 term. SIAM Activity Groups (SIAGs) provide a more focused forum for SIAM members interested in exploring one of the areas of applied mathematics, computational science, or applications.

Dr. Vijay Gadepally Receives AFCEA International’s 40 Under 40 Award

2021-07-29T00:00:00Z

Dr. Vijay Gadepally Receives AFCEA International’s 40 Under 40 Award Dr. Vijay Gadepally was selected as a recipient of the Armed Forces Communications and Electronics Association (AFCEA) International’s 40 Under 40 Award for his work in the establishment and continued growth of the Lincoln Laboratory Supercomputing Center, which hosts New England’s most powerful supercomputer.

MIT Lincoln Laboratory Supercomputing Center Innovations

2021-07-27T00:00:00Z

MIT Lincoln Laboratory Supercomputing Center Innovations The Lincoln Laboratory Supercomputing Center (LLSC) merges traditional HPC and Big Data technologies in an interactive on-demand parallel computing environment.By augmenting the processing power of desktop systems with high performance computational clusters, the LLSC enables researchers to develop and enhance algorithms for sensor data processing, high-fidelity simulations, and data science.

Supercomputing Center Celebrates Fifth Anniversary

2021-05-14T00:00:00Z

Supercomputing Center Celebrates Fifth Anniversary The Lincoln Laboratory Supercomputing Center (LLSC) celebrated a significant anniversary this year, marking five years of the center’s mission to enhance the computing power available to the Laboratory, MIT, and other researchers.

Cadets Collaboratively Intern Through the Air Force-MIT Artificial Intelligence Accelerator

2021-05-14T00:00:00Z

Cadets Collaboratively Intern Through the Air Force-MIT Artificial Intelligence Accelerator The Department of the Air Force-MIT Artificial Intelligence Accelerator (DAF-MIT AIA) is a collaboration between the Air and Space Forces and MIT to create new technology that will help the Air Force better complete their mission. The demand for interns on this program was fulfilled by ROTC cadets from Lincoln Laboratory and MIT.

Staff Member Selected as 2021 SIAM Fellow

2021-04-23T00:00:00Z

Staff Member Selected as 2021 SIAM Fellow The Society for Industrial and Applied Mathematics has selected Dr. Jeremy Kepner, Laboratory Fellow, Supercomputing Center, as an esteemed member of its 2021 class of SIAM Fellows.

LINEAR Is On the Watch for Potentially Hazardous Asteroids

2021-02-05T00:00:00Z

LINEAR Is On the Watch for Potentially Hazardous Asteroids NASA estimates that an asteroid the size of a car enters Earth’s atmosphere about once a year, creating a great fireball while burning up before reaching Earth’s surface; and roughly every 2,000 years, a football-stadium-sized meteoroid strikes Earth potentially causing significant damage. When will the next dangerous asteroid penetrate the atmosphere and seriously impact the Earth? Could that next asteroid be large enough to jeopardize civilization or the future of the human species? The Laboratory has been working since the late 1990s to help with the discovery and characterization of potentially hazardous asteroids. Laboratory researchers have found approximately one quarter of all known near-Earth objects (NEOs) that are at least 140 meters (460 feet) in size—large enough to have significant regional effects were they to impact the Earth.

Bug-Injecting System Helps to Advance the State-of-the-Art in Debugging Software

2021-03-05T00:00:00Z

Bug-Injecting System Helps to Advance the State-of-the-Art in Debugging Software Bug finding systems are used after developers have written code to try to identify mistakes they have made. If these systems find a bug, they can be fixed before code is deployed. Unfortunately, these systems fail to find many bugs, which is one of the reasons why new vulnerabilities and crashes still exist in computer programs today. The scarce documentation of known bugs and how those bugs manifest in a program made it impossible to measure the success of bug-finding tools. The Large-scale Automated Vulnerability Addition (LAVA) system enables evaluation of bug-finding systems. The LAVA system produces thousands of realistic bugs that are automatically injected into pre-existing program code. Once these bugs are injected, various vulnerability discovery techniques and software can be tested to see how many of the bugs are found and how many are missed. Over the last five years, LAVA has become the first widely used benchmark for evaluation of bug-finding systems.

MIT Lincoln Laboratory Supercomputing Center Brochure

2018-09-20T00:00:00Z

MIT Lincoln Laboratory Supercomputing Center Brochure The Lincoln Laboratory Supercomputing Center (LLSC) was established to better address supercomputing needs across all Laboratory missions, develop new supercomputing capabilities and technologies, and spawn even closer collaborations with MIT campus supercomputing initiatives. The center has a unique focus on interactive supercomputing for high-performance data analysis, and is located in an extremely ‘green’ computing center in Holyoke, Massachusetts, allowing our computers to run 100% carbon free.

Brainstorming energy-saving hacks on Satori, MIT’s new supercomputer

2020-02-11T00:00:00Z

Brainstorming energy-saving hacks on Satori, MIT’s new supercomputer Martineau, Kim Students participated in the Green AI Hackathon, co-sponsored by the MIT Research Computing Project and MIT-IBM Watson AI Lab, to explore methods for making artificial intelligence faster and more sustainable.

Wilkinson Prize goes to developers of flexible Julia programming language

2019-03-12T00:00:00Z

Wilkinson Prize goes to developers of flexible Julia programming language Ryan, Dorothy Programmers developing applications for fields as diverse as astronomy, economics, artificial intelligence, energy optimization, and medicine often found themselves creating software with languages that offered slow computation. But in this era of big data, dynamic, flexible, and easy-to-implement code is required for programmers to efficiently build high-performance software tools needed for intensive data analysis. Enter Julia, an open-source language for advanced technical computing and data science.

How Supercomputers Are Helping to Fight COVID-19

2020-03-23T00:00:00Z

How Supercomputers Are Helping to Fight COVID-19 Mandelbaum, Ryan F. A host of companies, including IBM, Microsoft, and Google, along with universities and national labs have teamed up to form the COVID-19 High Performance Computing (HPC) Consortium. This new partnership is designed to provide scientists with supercomputing resources as they figure out how to combat the coronavirus-caused disease known as COVID-19.

An algorithm with an eye for visibility helps pilots in Alaska

2019-06-17T00:00:00Z

An algorithm with an eye for visibility helps pilots in Alaska Foye, Kylie In remote areas of Alaska, pilots check current or forecasted weather conditions before they fly, but a lack of automated weather observation stations throughout the Alaskan bush makes it hard to know exactly what to expect. To help, the FAA recently installed 221 web cameras near runways and mountain passes. Pilots can look at the image feeds online to plan their route. Still, it’s difficult to go through what could be hundreds of images and estimate just how far one can see. Laboratory staff have been working with the FAA to turn these web cameras into visibility sensors and have developed an algorithm, called Visibility Estimation through Image Analytics (VEIA), that uses a camera’s image feed to automatically determine the area’s visibility. These estimates can then be shared among forecasters and with pilots online in real time.

Record-breaking DNA comparisons drive fast forensics

2019-06-17T00:00:00Z

Record-breaking DNA comparisons drive fast forensics DNA forensics is a powerful tool, yet it presents a computational scaling problem when it is improved and expanded for complex samples (those containing DNA from more than one individual) and kinship analysis. Laboratory staff developed an integrated web-based platform that provides expanded comparison capabilities without compromising speed or functionality. These new algorithms encode genetic markers as bits to allow for fast DNA comparisons in forensics.

Video and Imagery Dataset to Drive Public Safety Capabilities

2019-08-23T00:00:00Z

Video and Imagery Dataset to Drive Public Safety Capabilities Laboratory staff have been developing a computer vision dataset of operational and representative public safety scenarios. This dataset will enable technology development tailored to public safety scenarios, and includes operational images and videos from several organizations. They have labeled images so that machine learning algorithms can recognize a wide range of relevant public safety features in different environments. “The information within these images could improve various aspects of a response and recovery effort, such as damage assessment. Our dataset will enable the development of machine-learned analytics to prioritize and characterize images.”

MIT SuperCloud

2015-05-08T00:00:00Z

MIT SuperCloud The “big data” problem is pervasive across the DoD and is most commonly characterized by the three “Vs” of big data: volume, velocity, and variety. Now, increasingly, a fourth “V”, veracity (i.e., security), is becoming prominent. The MIT SuperCloud project has demonstrated significant quantitative impact on all of these areas. A key element of the MIT SuperCloud is its database management system, which allows unlimited instances of the National Security Agency (NSA)–developed Apache Accumulo database to be deployed on a project-by-project basis. This capability has moved Lincoln Laboratory to the forefront of the DoD big data community, as Accumulo becomes an increasingly widely used database for the U.S. intelligence community.

Establishment of the Lincoln Laboratory Supercomputing Center

2016-04-15T00:00:00Z

Establishment of the Lincoln Laboratory Supercomputing Center Lincoln Laboratory has been a world leader in interactive supercomputing since the 1950s. Recently, the Laboratory acknowledged the importance of the LLGrid world-class computing capability with the establishment of the Lincoln Laboratory Supercomputing Center (LLSC) on 1 April. “By establishing the LLSC, Lincoln Laboratory will be able to better address supercomputing needs across all Laboratory missions, develop new supercomputing capabilities and technologies, and spawn even closer collaborations with MIT campus supercomputing initiatives,” said Dr. Jeremy Kepner, Laboratory Fellow, and head of the Supercomputing Center.

A Tiny Organism with a Big Data Problem

2016-09-09T00:00:00Z

A Tiny Organism with a Big Data Problem Prochlorococcus is the smallest and most abundant photosynthetic organism on earth. Despite its tiny size, it’s an organism of global importance. In recent decades, researchers have sequenced the organisms’ genomes. Advances in sequencing technologies have generated massive databases of ocean genomic data from around the world. So while there is rich data available about Prochlorococcus, mining the value of this Big Data is difficult because it requires simultaneously analyzing various types of complex information. For the past six months, the team has worked with the Chisholm Lab at MIT to develop applications within their BigDAWG architecture to fit the specific needs of the lab.

Expanding Air Traffic Controllers’ View of Offshore Weather

2016-10-14T00:00:00Z

Expanding Air Traffic Controllers’ View of Offshore Weather Researchers at the Laboratory, working with the Federal Aviation Administration, have designed the Offshore Precipitation Capability (OPC) to address the lack of airspace situational awareness for aircraft traversing oceanic sectors of the National Airspace System (NAS).

Laboratory’s Supercomputing System Ranked Most Powerful in New England

2016-12-02T00:00:00Z

Laboratory’s Supercomputing System Ranked Most Powerful in New England The new computing system, TX-Green, at the Lincoln Laboratory Supercomputing Center (LLSC) has been named the most powerful supercomputer in New England and the 3rd most powerful at a United States university on the TOP500 list of the world’s 500 most powerful supercomputers.

Graph Processor Prototype

2017-02-10T00:00:00Z

Graph Processor Prototype In order to achieve significantly better graph computation performance, an advanced multiprocessor architecture has been developed that is optimized for analysis of large databases.

Staff Release Open Source Software for BigDAWG Polystore System

2017-04-14T00:00:00Z

Staff Release Open Source Software for BigDAWG Polystore System

Enabling Massive Computation and Resiliency in the Internet-of-Things Era

2017-06-16T00:00:00Z

Enabling Massive Computation and Resiliency in the Internet-of-Things Era The Internet of Things (IoT), an ever-growing network of physical devices connected to the Internet, brings a unique set of challenges to the Department of Defense (DoD). These challenges include the billions of connected devices, the tremendous diversity of the data being generated by these devices, and the varied defenses required to protect the applications. To address challenges in working with diverse datasets, Laboratory staff and university collaborators created the BigDAWG polystore system.

DataSToRM: Data Science and Technology Research Environment

2018-02-09T00:00:00Z

DataSToRM: Data Science and Technology Research Environment By analyzing graphs using specialized algorithms, complex relationships and deeper insight can be extracted from the raw information. In the last several years, the Laboratory has developed the Graph Processor, which has a unique hardware architecture that provides 100 to 1000 times better processing performance for analyzing large graph datasets. The Data Science and Technology Research Environment (DataSToRM) program is developing a software environment and algorithms to take advantage of the Graph Processor’s capabilities.

Winds Forecast Rapid Prototype

2018-06-08T00:00:00Z

Winds Forecast Rapid Prototype Wind causes significant problems for aviation, yet the Federal Aviation Administration has limited tools for forecasting its impact. MIT Lincoln Laboratory is developing technology to generate improved wind forecasts through the use of data fusion and machine learning.

New Textbook Applies Mathematics to the Management of Big Data

2018-08-10T00:00:00Z

New Textbook Applies Mathematics to the Management of Big Data Mathematics of Big Data is the first book to present the common mathematical foundations of big data analysis across a range of applications and technologies.

Lidar Scans Over the Carolinas Accelerate Hurricane Recovery

2019-01-11T00:00:00Z

Lidar Scans Over the Carolinas Accelerate Hurricane Recovery The Federal Emergency Management Agency (FEMA) called upon MIT Lincoln Laboratory to use its state-of-the-art light detection and ranging (lidar) system to image the destruction from hurricane Florence. A high-resolution 3D model of the scanned area depicts the heights of structures and landscape features. Laboratory analysts can then process this data to glean information that helps FEMA focus their recovery efforts—for example, by estimating the number of collapsed houses in an area, the volume of debris piles, and the reach of flood waters.

Supercomputers Can Spot Cyber Threats

2019-01-18T00:00:00Z

Supercomputers Can Spot Cyber Threats McGovern, Anne Lincoln Laboratory researchers have developed a technique to compress hours of internet traffic into a bundle that can be analyzed for suspicious behavior.

Creating Synthetic Radar Imagery Using Convolutional Neural Networks

2019-02-15T00:00:00Z

Creating Synthetic Radar Imagery Using Convolutional Neural Networks Weather radar can track the location and intensity of storms and is useful for managing transportation around hazardous weather. Air traffic controllers, for example, rely on weather radar to track storms that could impact aircraft and flight schedules. Although land-based radar is sufficient to cover most continental air space, many offshore and oceanic controllers do not have sufficient access to the weather information that they need for proper air traffic management. Researchers at MIT Lincoln Laboratory developed the Offshore Precipitation Capability (OPC), a system that creates a radar-like depiction of precipitation — known as synthetic radar — by combining data from multiple nonradar sources, and implementing convolutional neural networks into the OPC system as a means of combining the nonradar data sources to create synthetic radar.

MIT and U.S. Air Force Sign Agreement to Launch AI Accelerator

2019-07-12T00:00:00Z

MIT and U.S. Air Force Sign Agreement to Launch AI Accelerator MIT and the U.S. Air Force have signed an agreement to launch a new program designed to make fundamental advances in artificial intelligence that could improve Air Force operations while also addressing broader societal needs. The new program will focus on rapid deployment of artificial intelligence innovations in operations, disaster response, and medical readiness.

Enabling the Foundations of AI: Data, Computation, and Algorithms

2019-11-01T00:00:00Z

Enabling the Foundations of AI: Data, Computation, and Algorithms MIT Lincoln Laboratory staff in the Lincoln Laboratory Supercomputing Center are developing tools to address challenges in data management and algorithmic techniques for novel neural network architectures to enable rapid prototyping of AI.

LLx Project Seeks to Improve Online Hands-On Learning

2019-12-06T00:00:00Z

LLx Project Seeks to Improve Online Hands-On Learning Online learning has been around for decades, but educators still struggle to adapt certain hands-on subjects to the web. At MIT Lincoln Laboratory, a team of researchers has been working on ways to close this practical learning gap through the Lincoln Laboratory Online Courses (LLx) project. The project was born from a desire to adapt the Laboratory’s internal course offerings for a wider audience. Its goal is to identify best practices for online hands-on learning in order to provide unique Laboratory classes to sponsors, students, and the general public as self-paced, massively open online courses (MOOCs).

Cybersecurity Phenomenology Exploration

2020-03-20T00:00:00Z

Cybersecurity Phenomenology Exploration

New Algorithm Uses Supercomputing to Combat Cyber Attacks

2020-05-15T00:00:00Z

New Algorithm Uses Supercomputing to Combat Cyber Attacks Sophisticated cyber attacks are on the rise. Early techniques for cyber attacks, such as guessing passwords manually, have evolved throughout the years—from session hijacking to ransomware and beyond. In order to manage this tricky cyber attack landscape, analysts need automated tools that can accurately detect and classify threats. Researchers are using machine learning to characterize anomalous behavior within a cyber network.

High-Fidelity Multi-Physics Simulations for Hypersonics

2020-06-12T00:00:00Z

High-Fidelity Multi-Physics Simulations for Hypersonics

The Laboratory’s New AI Supercomputer is the Most Powerful at any University

2019-07-19T00:00:00Z

The Laboratory’s New AI Supercomputer is the Most Powerful at any University The new TX-GAIA computing system at the Lincoln Laboratory Supercomputing Center (LLSC) has been ranked as the most powerful artificial intelligence (AI) supercomputer at any university in the world. The ranking comes from TOP500, which publishes a list of the top supercomputers in various categories biannually.

Staff Build Robust Algorithms to Strengthen Machine Learning Methods

2020-06-26T00:00:00Z

Staff Build Robust Algorithms to Strengthen Machine Learning Methods “Our ultimate goal is to provide the researchers and stakeholders of our community with a set of robustness tools, techniques, and best practices so that they can embrace the great promise of machine learning technology with the confidence that they can meet the safety and security demands that are specific to the national security domain,”

Mission-ready Reinforcement Learning

2020-08-28T00:00:00Z

Mission-ready Reinforcement Learning The Mission-ready reinforcement learning (MeRLin) program is looking to solve complex planning and coordination problems across a range of Department of Defense mission areas. MeRLin is focusing on developing and training Deep reinforcement learning (DRL) algorithms capable of maintaining performance on complex tasks with human allies.