Summer Internship at Institut des Systèmes Intelligents et de Robotique, Sorbonne Université
The goal of this internship is to implement different models of rational and irrational humans.
- ToyEnvironment: First implementation of a Boltzmann-noisy-rationnal agent trained by Q-Learning on a simple homemade grid environment
- GymMazeEnvironment : Boltzmann-noisy-rationnal and irrationnal biases on OpenAi's GymMaze environment (https://github.com/MattChanTK/gym-maze)