This project provides code to reproduce the AIM algorithm used in Approximate information maximization for bandit games.
The code allows to simulate bandit game with Bernoulli and Gaussian reward but isn't meant to reproduce the figures given in the main paper.
\ No newline at end of file
This project provides code to reproduce the AIM algorithm used in Aistats 2025 paper "Approximate information maximization for bandit games".
This code enables the simulation of a bandit game with Bernoulli and Gaussian rewards but is not intended to replicate the figures presented in the main paper.
