A5000327528- Infrastructure Resilience and Vulnerability Analysis
- Multi-Agent Systems and Negotiation
- Reinforcement Learning in Robotics
- AI-based Problem Solving and Planning
- Constraint Satisfaction and Optimization
- Evacuation and Crowd Dynamics
- Game Theory and Applications
- Logic, Reasoning, and Knowledge
- Advanced Bandit Algorithms Research
- Information and Cyber Security
- Network Security and Intrusion Detection
- Military Defense Systems Analysis
- Complex Network Analysis Techniques
- Auction Theory and Applications
- Smart Grid Energy Management
- Opinion Dynamics and Social Influence
- Smart Grid Security and Resilience
- Homelessness and Social Issues
- COVID-19 epidemiological studies
- Crime Patterns and Interventions
- Optimization and Search Problems
- Terrorism, Counterterrorism, and Political Violence
- Bayesian Modeling and Causal Inference
- Artificial Intelligence in Games
- Machine Learning and Algorithms
Harvard University Press
2019-2025
Harvard University
2019-2024
Google (United States)
2020-2024
University of Southern California
2013-2022
Southern California University for Professional Studies
2011-2021
Carnegie Mellon University
1988-2020
Boston Children's Hospital
2020
National Bureau of Economic Research
2020
Data & Society Research Institute
2020
IIT@MIT
2020
The COVID-19 pandemic has created a public health crisis. Because SARS-CoV-2 can spread from individuals with presymptomatic, symptomatic, and asymptomatic infections, the reopening of societies control virus will be facilitated by robust population screening, for which testing often central. After infection, undergo period incubation during viral titers are too low to detect, followed exponential growth, leading peak load infectiousness ending declining clearance. Given pattern kinetics, we...
In a class of games known as Stackelberg games, one agent (the leader) must commit to strategy that can be observed by the other follower or adversary) before adversary chooses its own strategy. We consider Bayesian in which leader is uncertain about types it may face. Such are important security domains, where, for example, (leader) patrolling certain areas, and robber (follower) has chance observe this over time choosing where attack. This paper presents an efficient exact algorithm...
Security at major locations of economic or political importance is a key concern around the world, particularly given threat terrorism. Limited security resources prevent full coverage all times, which allows adversaries to observe and exploit patterns in selective patrolling monitoring, e.g. they can plan an attack avoiding existing patrols. Hence, randomized monitoring important, but randomization must provide distinct weights different actions based on their complex costs benefits. To...
Abstract The COVID-19 pandemic has created a public health crisis. Because SARS-CoV-2 can spread from individuals with pre-symptomatic, symptomatic, and asymptomatic infections, the re-opening of societies control virus will be facilitated by robust surveillance, for which testing often central. After infection, undergo period incubation during viral titers are usually too low to detect, followed an exponential growth, leading peak load infectiousness, ending declining levels clearance....
Poaching is a serious threat to the conservation of key species and whole ecosystems. While conducting foot patrols most commonly used approach in many countries prevent poaching, such often do not make best use limited patrolling resources. To remedy this situation, prior work introduced novel emerging application called PAWS (Protection Assistant for Wildlife Security); was proposed as game-theoretic (“security games”) decision aid optimize resources.This paper reports on PAWS’s...
Creating impact in real-world settings requires artificial intelligence techniques to span the full pipeline from data, predictive models, decisions. These components are typically approached separately: a machine learning model is first trained via measure of accuracy, and then its predictions used as input into an optimization algorithm which produces decision. However, loss function train may easily be misaligned with end goal, make best decisions possible. Hand-tuning align difficult...
In September 2016, Stanford's "One Hundred Year Study on Artificial Intelligence" project (AI100) issued the first report of its planned long-term periodic assessment artificial intelligence (AI) and impact society. It was written by a panel 17 study authors, each whom is deeply rooted in AI research, chaired Peter Stone University Texas at Austin. The report, entitled "Artificial Intelligence Life 2030," examines eight domains typical urban settings which likely to have over coming years:...
Interactive simulation environments constitute one of today's promising emerging technologies, with applications in areas such as education, manufacturing, entertainment, and training. These are also rich domains for building investigating intelligent automated agents, requirements the integration a variety agent capabilities but without costs demands low-level perceptual processing or robotic control. Our project is aimed at developing humanlike, agents that can interact each other, well...
Distributed Constraint Optimization (DCOP) is an elegant formalism relevant to many areas in multiagent systems, yet complete algorithms have not been pursued for real world applications due perceived complexity. To capably capture a rich class of complex problem domains, we introduce the Multi-Event Scheduling (DiMES) framework and design congruent DCOP formulations with binary constraints which are proven yield optimal solution. approach real-world efficiency requirements, obtain immense...
The increasing threat of terrorism makes security at major locations economic or political importance a concern. Limited resources prevent complete coverage, allowing adversaries to observe and exploit patterns in patrolling monitoring, enabling them plan attacks that avoid existing patrols. use randomized policies are more difficult for predict can counter their surveillance capabilities. We describe two applications, ARMOR IRIS, assist forces randomizing operations. These applications...
There has been significant recent interest in game-theoretic approaches to security, with much of the research focused on utilizing leader-follower Stackelberg game model. Among major applications are ARMOR program deployed at LAX Airport and IRIS use by US Federal Air Marshals (FAMS). The foundational assumption for using games is that security forces (leaders), acting first, commit a randomized strategy; while their adversaries (followers) choose best response after surveillance this...
While three deployed applications of game theory for security have recently been reported at AAMAS [12], we as a community remain in the early stages these deployments; there is continuing need to understand core principles innovative theory. Towards that end, this paper presents PROTECT, game-theoretic system by United States Coast Guard (USCG) port Boston scheduling their patrols. USCG has termed deployment PROTECT success, and efforts are underway test it New York, with potential...
We present a new polynomial-space algorithm, called Adopt, for distributed constraint optimization (DCOP). DCOP is able to model large class of collaboration problems in multi agent systems where solution within given quality parameters must be found. Existing methods are not provide theoretical guarantees on global while operating both efficiently and asynchronously. Adopt guaranteed find an optimal solution, or user-specified distance from the optimal, allowing agents execute...