Charting the Future of Artificial Intelligence
Applied research and mission innovation inspire tomorrow
Booz Allen’s AI leadership is driven by our mission focus and sustained commitment to applied research and scientific discovery. Our dedication to these priorities keeps us at the forefront of technological advancement, allowing us to anticipate and help our clients capitalize on emerging trends.
Our approach to AI research often begins with an opportunity to solve a difficult problem: an algorithm too slow to scale for the government’s data, a mission need that has no solution, or the physical restrictions on running applications on a truck in a disconnected environment. By focusing on the hardest problems our clients face in fulfilling their missions and serving the public, we inspire and empower a diverse set of researchers to bring their unique talents fully to bear in solving the nation’s most critical challenges.
Latest Thinking
Fostering a Research-Driven AI Culture
Explore how Booz Allen places research and innovation at the forefront of our AI culture:
Booz Allen experts produce world-class AI research on topics of national importance.
As the number-one provider of AI services to the nation, Booz Allen is a magnet for some of the world’s best AI researchers, who seek to transform their technical vision into real-world innovation and tangible mission outcomes. Our researchers focus their tremendous talent and expertise on solving specific challenges of national importance, such as improving healthcare, combating climate change, and safeguarding the nation. As evidence of their rigor and impact, Booz Allen researchers have published over 100 papers on emerging AI topics in peer-reviewed conferences and journals like Neural Information Processing Systems (NeurIPS) and the Conference on Computer Vision and Pattern Recognition (CVPR).
A firmwide network links Booz Allen with the global academic community.
We anticipate mission needs and partner with forward-thinking industry, startup, and academic leaders focused on innovation, safety, security, and progress. For example, we actively collaborate with computer science labs and math departments at Harvard University, Syracuse University, the Montreal-based Mila Institute, and other organizations. And our university-wide master collaboration agreement with the University of Maryland, Baltimore County, lets Booz Allen practitioners work on interdisciplinary projects with any professor from any department.
We maintain world-class technical capabilities that open new avenues for research.
We invest in advanced computing infrastructure to support specific client projects and to expand internal research opportunities. For example, our teams use a state-of-the-art graphics processing unit (GPU) cluster to explore their areas of research focus across AI-related disciplines, from AI security and malware detection to computer vision, generative AI, and more. Investing in AI also means investing in data that enables our teams to lead in producing new research and solving real-world problems in cancer pain detection, cybersecurity, and robotics.
We provide mentoring and ongoing opportunities for emerging researchers to explore cutting-edge AI technologies.
A Booz Allen initiative provides computer science and math graduates from universities, including MIT and Harvard, with opportunities to author papers and conduct other research activities. Participants build on their prior undergraduate and graduate coursework and their university research projects to rapidly hone skills and publish findings. We also work with the University of Notre Dame’s ESTEEM program to help members gain learning experiences in entrepreneurship and innovation that build upon their advanced science and engineering backgrounds.
Diversity across our sector-leading team drives game-changing innovations.
Exceptional solutions never come from doing the same thing the way it’s always been done. That’s why we recruit from a diverse set of backgrounds, skills, and potential to develop a breadth of internal knowledge. From linguists to machine learning engineers, university interns to Ph.D.s, fresh thoughts and novel perspectives allow us to develop technical strategies that bridge the gap between today’s need and tomorrow’s solution. Whether it’s serving the public more efficiently, turning raw data into intelligence insights, strengthening public health, or thwarting adversaries in cyberspace and on the battlefield, our diverse teams work collaboratively to drive research with real-world impact.
Review over 100 research papers from Booz Allen experts and explore career opportunities to learn more about how we engineer the next AI solutions based on peer-reviewed research and practical innovation.
Artificial Intelligence
Our AI research covers a wide gamut of low-level performance optimization, novel deep learning, reproducibility, and other issues informed by the needs of real-world deployments.
-
A Siren Song of Open Source Reproducibility - Outstanding Paper Award
-
Image Translation of Faces from Visible to Thermal Modalities using Auxiliary Domain and Sensor Labels
-
Exact Acceleration of K-Means++ and K-Means||
-
Multi-Task Graphs with Autoencoders
-
Accounting for Variance in Machine Learning Benchmarks
-
Research Reproducibility as a Survival Analysis
-
Bringing UMAP Closer to the Speed of Light with GPU Acceleration
-
Sampling Approach Matters: Active Learning for Robotic Language Acquisition
-
Modeling Watershed Nutrient Concentrations with AutoML
-
Robust Design of Deep Neural Networks against Adversarial Attacks based on Lyapunov Theory,
-
A New Burrows Wheeler Transform Markov Distance
-
A Step Toward Quantifying Independently Reproducible Machine Learning Research
-
PyLZJD: An Easy to Use Tool for Machine Learning
-
Hash-Grams On Many-Cores and Skewed Distributions
-
A Manifold Alignment Approach to Grounded Language Learning
-
Heterogeneous Relational Kernel Learning
-
Barrage of Random Transforms for Adversarially Robust Defense
-
Adversarial Attacks, Regression, and Numerical Stability Regularization
-
Practical Cross-modal Manifold Alignment for Robotic Grounded Language Learning
-
Learning to Make Predictions on Graphs with Autoencoders
-
Growing and Retaining AI Talent for the United States Government
-
Neural Fingerprint Enhancement
-
Hash-Grams: Faster N-Gram Features for Classification and Malware Detection
-
Linear Models with Many Cores: A Stochastic Atomic Update Scheme
Health and Responsible AI
Our research in fairness and ethics is largely driven by our experience in health and medical applications, working to tackle both technical and policy issues.
-
Wastewater Surveillance Pilot at U.S. Military Installations Sanjak, J. S., McAuley, E. M., Raybern, J., Pinkham, R., Tarnowski, J., Miko, N., ... & Maier, E. J. 2024
-
SARS-CoV-2 Genome-Based Severity Predictions Correspond to Lower qPCR Values and Higher Viral Load Martin Skarzynski, Erin M. McAuley, Ezekiel J. Maier, Anthony C. Fries, Jameson D. Voss, and Richard R. Chapleaua 2022
-
Finding Potential Adverse Events in the Unstructured Text of Electronic Health Care Records: Development of the Shakespeare Method Bright R, Rankin S, Dowdy K, Blok S, Bright S, Palmer L. JMIRx Med 2021;2(3):e27017
-
Towards the Use of Neural Networks for Influenza Prediction at Multiple Spatial Resolutions Emily L Aiken, Andre T Nguyen, Cecile Viboud, Mauricio Santillana. Science Advances
-
Estimating the Cumulative Incidence of COVID-19 in the United States Using Four Complementary Approaches Fred S Lu, Andre T Nguyen, Nicholas B Link, Mathieu Molina, Jessica T Davis, Matteo Chinazzi, Xinyue Xiong, Alessandro Vespignani, Marc Lipsitch, Mauricio Santillana. PLOS Computational Biology
-
An Early Warning Approach to Monitor COVID-19 Activity with Multiple Digital Traces in Near Real-Time Nicole E Kogan, Leonardo Clemente, Parker Liautaud, Justin Kaashoek, Nicholas B Link, Andre T Nguyen, Fred S Lu, Peter Huybers, Bernd Resch, Clemens Havas, Andreas Petutschnig, Jessica Davis, Matteo Chinazzi, Backtosch Mustafa, William P Hanage, Alessandro Vespignani, Mauricio Santillana.. Science Advances
-
Real-time Estimation of Disease Activity in Emerging Outbreaks Using Internet Search Information Emily L Aiken, Sarah F McGough, Maimuna S Majumder, Gal Wachtel, Andre T Nguyen, Cecile Viboud, Mauricio Santillana.. PLOS Computational Biology
-
Flexible and Adaptive Fairness-aware Learning in Non-stationary Data Streams Wenbin Zhang; Mingli Zhang; Ji Zhang; Zhen Liu; Zhiyuan Chen; Jianwu Wang; Edward Raff; Enza Messina.. Proceedings of the 32th International Conference on Tools with Artificial Intelligence (ICTAI)
-
The Use of AI for Thermal Emotion Recognition: A Review of Problems and Limitations in Standard Design and Data Ordun, C.; Raff, E.; Purushotham, S. AAAI FSS-20 AI in Government and Public Sector Applications
-
COVID-19 Kaggle Literature Organization Eren, M. E., Solovyev, N., Raff, E., Nicholas, C., & Johnson, B. Proceedings of the ACM Symposium on Document Engineering
-
Exploratory Analysis of COVID-19 Tweets Using Topic Modeling, UMAP, and DiGraphs Ordun, C.; Purushotham, S.; Raff, E. epiDAMIK 2020: 3rd EpiDAMIK ACM SIGKDD International Workshop on Epidemiology Meets Data Mining and Knowledge Discovery
-
Recommendations for the Use of Social Media in Pharmacovigilance: Lessons from IMI WEB-RADR van Stekelenborg J, Ellenius J, Maskell S, Bergvall T, Caster O, Dasgupta N, Dietrich J, Gama S, Lewis D, Newbould V, Brosch S, Pierce CE, Powell G, Ptaszyńska-Neophytou A, Wiśniewski AFZ, Tregunno P, Norén GN, Pirmohamed M.. Drug Safety
-
Early Outcomes and Complications of Obese Patients Undergoing Shoulder Arthroplasty: a Meta-Analysis Klein, A.; Jauregui, J.; Raff, E.; Gilotra, M. Journal of Clinical Orthopaedics and Trauma
-
Recommendations on the Use of Mobile Applications for the Collection and Communication of Pharmaceutical Product Safety Information: Lessons from IMI WEB-RADR Pierce CE, de Vries ST, Bodin-Parssinen S, Härmark L, Tregunno P, Lewis DJ, Maskell S, Van Eemeren R, Ptaszynska-Neophytou A, Newbould V, Dasgupta N, Wisniewski AFZ, Gama S, Mol PGM. Drug Safety
-
Towards the Use of Neural Networks for Influenza Prediction at Multiple Spatial Resolutions Emily L Aiken, Andre T Nguyen, Mauricio Santillana. NeurIPS 2019 ML4H: Machine Learning for Health Workshop
-
Exploratory Analysis of Covid-19 Tweets using Topic Modeling, UMAP, and DiGraphs Catherine Ordun, Sanjay Purushotham, Edward Raff. 2020
-
Intelligent Sight and Sound: A Chronic Cancer Facial Pain Dataset Catherine Ordun, Alexandra Cha, Edward Raff, Byron Gaskin, Alexander Hanson, Mason Rule, Sanjay Purushotham, James Gulley. 2021
-
Bias Amplification in Artificial intelligence Systems Lloyd, Kirsten. AAAI Fall Symposium 2018. Presented at AAAI FSS-18: Artificial Intelligence in Government and Public Sector, Arlington, Virginia, USA
-
Improved Automatic Pharmacovigilance: An Enhancement to the MedWatcher Social System for Monitoring Adverse Events Andre T Nguyen, Julia Lien, Edward Raff, Sumiko R Mekaru. Epidemiology Meets Data Mining and Knowledge Discovery Workshop at KDD 2019
-
Gradient Reversal Against Discrimination Edward Raff, Jared Sylvester. In Proceedings of the 5th Workshop on Fairness, Accountability and Transparency in Machine Learning. Stockholm, Sweden, 2018
-
Gradient Reversal Against Discrimination: A Fair Neural Network Learning Approach Edward Raff, Jared Sylvester. IEEE Conference on Data Science and Advanced Analytics (DSAA)
-
What About Applied Fairness? Sylvester, Jared, and Edward Raff. In Machine Learning: The Debates (ML-D) Organized as Part of the Federated AI Meeting (FAIM 2018), 2018
-
Evaluation of Facebook and Twitter Monitoring to Detect Safety Signals for Medical Products: An Analysis of Recent FDA Safety Alerts Pierce CE, Bouri K, Pamer C, Proestel S, Rodriguez HW, Van Le H, Freifeld CC, Brownstein JS, Walderhaug M, Edwards IR, Dasgupta N. Drug Safety, 2017
-
Combining Participatory Influenza Surveillance with Modeling and Forecasting: Three Alternative Approaches John S Brownstein, Shuyu Chu, Achla Marathe, Madhav V Marathe, Andre T Nguyen, Daniela Paolotti, Nicola Perra, Daniela Perrotta, Mauricio Santillana, Samarth Swarup, Michele Tizzoni, Alessandro Vespignani, Anil Kumar S Vullikanti, Mandy L Wilson, Qian Zhang. JMIR Public Health and Surveillance, 2017
-
Social Media Listening for Routine Post-Marketing Safety Surveillance Powell GE, Seifert HA, Reblin T, Burstein PJ, Blowers J, Menius JA, Painter JL, Thomas M, Pierce CE, Rodriguez HW, Brownstein JS, Freifeld CC, Bell HG, Dasgupta N.. Drug Safety, 2016
-
Sustained Reduction of Diversion and Abuse After Introduction of an Abuse Deterrent Formulation of Extended Release Oxycodone Severtson SG, Ellis MS, Kurtz SP, Rosenblum A, Cicero TJ, Parrino MW, Gilbert MK, Buttram ME, Dasgupta N, BucherBartelson B, Green JL, Dart RC. Drug and Alcohol Dependence, 2016
-
Diversion and Illicit Sale of Extended Release Tapentadol in the United States Richard C. Dart, MD, PhD, Hilary L. Surratt, PhD, Marie-Claire Le Lait, MS, Yami Stivers, BA, Vikhyat S. Bebarta, MD, Clark C. Freifeld, PhD, John S. Brownstein, PhD, John J. Burke, AS, Steven P. Kurtz, PhD, Nabarun Dasgupta, PhD . Pain Med., 2016
-
Increasing Patient Engagement in Pharmacovigilance Through Online Community Outreach and Mobile Reporting Applications: An Analysis of Adverse Event Reporting for the Essure Device in the US. Bahk CY, Goshgarian M, Donahue K, Freifeld CC, Menone CM, Pierce CE, Rodriguez H, Brownstein JS, Furberg R, Dasgupta N.. Pharmaceutical Medicine, 2015
-
Crowdsourcing Black Market Prices for Prescription Opioids Dasgupta N, Freifeld C, Brownstein JS, Menone CM, Surratt HL, Poppish L, Green JL, Lavonas EJ, Dart RC. Journal of Medical Internet Research, 2013
-
Dr. AI, Where Did You Get Your Degree? Raff, Edward, Shannon Lantzy, and Ezekiel Maier. Artificial Intelligence in Health (pp. 76-83)
-
Dr. AI, Where Did You Get Your Degree? Raff, Edward, Shannon Lantzy, and Ezekiel Maier. Proceedings of the First Joint Workshop on AI in Health Organized as Part of the Federated AI Meeting (FAIM 2018) : 204–7
Security and Privacy
Our cybersecurity practice has shown that many off-the-shelf and best practices in machine learning and deep learning don’t translate to cybersecurity data. Because of this, we often develop new methods to meet the larger scale, non-independent, and identically distributed nature of real data. Many times, the low-latency and low-power requirements of being deployed on a network means custom code is needed to satisfy real-time requirements.
-
Adversarial Transfer Attacks With Unknown Data and Class Overlap Luke Richards; Andre Nguyen; Ryan Capps; Steven Forsyth; Cynthia Matuszek; Edward Raff. AISec
-
A Framework for Cluster and Classifier Evaluation in the Absence of Reference Labels Robert Joyce; Edward Raff; Charles Nicholas. AISec
-
Leveraging Uncertainty for Improved Static Malware Detection Under Extreme False Positive Constraints Andre Nguyen; Edward Raff; Charles Nicholas; James Holt. First International Workshop on Adaptive Cyber Defense
-
Classifying Sequences of Extreme Length with Constant Memory Applied to Malware Detection Edward Raff; William Fleshman; Richard Zak; Hyrum Anderson; Bobby Filar; Mark McLean. AAAI 2021
-
A Survey of Machine Learning Methods and Challenges for Windows Malware Classification Edward Raff, Charles Nicholas. ML-RSA @ NeurIPS
-
Getting Passive Aggressive About False Positives: Patching Deployed Malware Detectors Raff, E.; Holt, J.; Filar, B.. IEEE International Conference on Data Mining Workshop (ICDM) on Deep Learning for Cyber Threat Intelligence (DL-CTI)
-
Automatic Yara Rule Generation Using Biclustering Edward Raff; Richard Zak; Gary Lopez Munoz; William Fleming; Hyrum S. Anderson; Bobby Filar; Charles Nicholas; James Holt. 13th ACM Workshop on Artificial Intelligence and Security (AISec’20)
-
Would a File by Any Other Name Seem as Malicious? Nguyen, A.; Raff, E.; Sant-Miller, A. IEEE Big Data
-
KiloGrams: Very Large N-Grams for Malware Classification Raff, Edward; Fleming, W.; Zak, Richard; Anderson, H.; Finlayson, B.; Nicholas, C.; McLean, M. KDD Workshop on Learning and Mining for Cybersecurity (LEMINCS)
-
Non-Negative Networks Against Adversarial Attacks Fleshman, William, Edward Raff, Jared Sylvester, Steven Forsyth, and Mark McLean. AAAI Workshop on Artificial Intelligence for Cyber Security (AICS)
-
Lempel-Ziv Jaccard Distance, An Effective Alternative to Ssdeep and Sdhash Edward Raff, Charles Nicholas. Digital Investigation
-
Static Malware Detection & Subterfuge: Quantifying the Robustness of Machine Learning and Current Anti-Virus Fleshman, William, Edward Raff, Richard Zak, Mark McLean, and Charles Nicholas. AAAI Fall Symposium, 2018.
-
Malware Detection by Eating a Whole EXE Edward Raff, Jon Barker, Jared Sylvester, Robert Brandon, Bryan Catanzaro, Charles Nicholas. AAAI Workshop on Artificial Intelligence for Cyber Security
-
An Alternative to NCD for Large Sequences, Lempel-Ziv Jaccard Distance Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
-
What Can N-Grams Learn for Malware Detection? Richard Zak, Edward Raff, Charles Nicholas. 2017 12th International Conference on Malicious and Unwanted Software (MALWARE) (pp. 109–118)
-
Malware Classification and Class Imbalance via Stochastic Hashed LZJD Edward Raff, Charles Nicholas. - Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security (pp. 111–120) 2017
-
Learning the PE Header, Malware Detection with Minimal Domain Knowledge Edward Raff, Jared Sylvester, Charles Nicholas. Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security (2017) 121-132
-
An Investigation of Byte N-Gram Features for Malware Classification Edward Raff, Richard Zak, Russell Cox, Jared Sylvester, Paul Yacci, Rebecca Ward, Anna Tracy, Mark McLean, Charles Nicholas. Journal of Computer Virology and Hacking Techniques
Quantum Information Sciences
Quantum technologies are unlocking powerful opportunities and threat vectors across all missions. Our quantum experts work closely with our mission partners on research to better understand, advance, and unlock these opportunities while also working aggressively to counter the cybersecurity threats these technologies may pose.
-
Optimal Two-Qubit Circuits for Universal Fault-Tolerant Quantum Computation
-
Effective Gaps Are Not Effective: Quasipolynomial Classical Simulation of Obstructed Stoquastic Hamiltonians
-
Spectrally Reconfigurable Quantum Emitters Enabled by Optimized Fast Modulation
-
A Nonlinear Regression for Characterizing Two-Qubit Processors
Meet Our Researchers
Our researchers create a culture to empower engineers and data scientists to innovate while we continue to move AI forward into the future.
