A custom MARL (multi-agent reinforcement learning) environment where multiple agents trade against one another (self-play) in a zero-sum continuous double auction. Ray [RLlib] is used for training.

A 2D JavaFX implementation of one of the most popular and old board games. The AI player's moves are calculated using the expectiminimax algorithm.

An analysis library for zero-sum games, written in Rust

Demonstration of the concept of convergence for the Fictitious Play algorithm, simulating a simple zero-sum game over multiple iterations.

Simulating the game theory behind Linaia-Agon by Iannis Xenakis

Psychopy implementation of the article : "Neural computations underpinning the strategic management of influence in advice giving" by Uri Hertz & BB

Desktop Reversi game made in Kotlin and Swing. The project focuses on implementing and exploring AI algorithms for zero-sum games, including MinMax, Alpha-Beta pruning, and various heuristics.

Some applications of optimization using linear, binary, and integer programming.

Zero Sum Budgeting Software

Framework, boilerplate, and examples for zerosum algorithms like minimax.

A tic tac toe game with an AI that you just can't beat

While working on my joint paper 'Davenport constant for finite abelian groups with higher rank' (arXiv:2402.09999 [math.NT]) with Dr. Eshita Mazumdar, I used these R programs to get an idea initially how our calculated bounds for the r-wise Davenport constant were behaving as we considered more complicated and bigger group structures.

my favorite projects

A minimax algorithm to solve a two-player connection board game