A research platform to develop automated security policies using quantitative methods, e.g., optimal control, computational game theory, reinforcement learning, optimization, evolutionary methods, and causal inference.

Implementing different learning algorithms and analyzing their performance in a Markov game model called the Soccer Game

Correct-by-synthesis reinforcement learning with temporal logic constraints (CoRL)

Board-and-card games are those which involve higher level of uncertainty as it includes the probability of getting the right card and the moves made by other players. We look to model such games as Markov Games and find an optimal policy through the Minimax – Q algorithm. This will also be a test for the Minimax – Q algorithm to check how it per…

Deterministic hex-grid soccer environment with two adversarial agents. Implements Q-Learning, Minimax-Q (via LP), and Belief-Q with online belief updates; trains in SE2G/SE6G to reduce state space and evaluates behaviors in the full environment with comprehensive visualizations.