Current Project Areas


Research Overview

The Teamcore group's roots are in Artificial Intelligence (AI). Our research in the past decade and longer has focused on "AI for social good", covering topics such as but not limited to: (i) AI for protection of forests, fish and wildlife; (ii) AI for public safety and security; (iii) AI for assisting low-resource sections of society. We focus on fundamental research problems in computational game theory, machine learning, automated planning, intelligent agents and multiagent interactions that are driven by these topics, ensuring a virtuous cycle of research and real-world applications. Examples of research driven by real-world problems include challenges of learning and handling bounded rationality models of human (adversary) behavior from real-world data, solving massive scale games uncertainty, algorithms that enable influence maximization in social networks under dynamism and uncertainty.

Our research is in real world use in the form of decision aids to assist Governmental and Non-Governmental Organizations in protecting wildlife, fisheries, and forests, as well as in assisting in public safety and security at airports or ports, or on trains and in university campuses. This includes the PROTECT system deployed by the US Coast Guard in a number of ports nationwide for planning effective randomized patrols, IRIS by the Federal Air Marshals service, the PAWS system for wildlife protection used by different NGOs; with active efforts to build decision aids for assisting health information dissemination in homeless shelters.




 

Previous Projects

 

maaf_raven

    Multiagent Systems Research for Sustainability

maaf_raven

    TRUSTS

maaf_raven

    PROTECT(Continues to be in use by the US Coast Guard since 2011)

maaf_raven

    ARMOR-IRIS(Continues to be in use by FAMS since October 2009)

maaf_raven

    ARMOR-LAX (Continues to be in use at LAX)

maaf_raven

    ARMOR-GUARDS

maaf_raven

    Planning and Coordination for Multiagent Teams

maaf_raven

    Learning Agents and Activity Detection

maaf_raven

    Multiagent Based Simulations: Cancer Modeling and Markov Decision Problems

    DCOP team formation for disaster response

    Disaster rescue simulations

 

    COORDINATORS

    Agents facilitated human organizations (CALO)

    Teamcore and Electric Elves  

 

 

    Distributed Resource Allocation (Dynamite) 

 

 

  

     Large Area Security