The Hierarchical Intrinsically Motivated Agent (HIMA) is an algorithm that is intended to exhibit an adaptive goal-directed behavior using neurophysiological models of the neocortex, basal ganglia, and thalamus.

[DSN 2024] Toward Evaluating Robustness of Reinforcement Learning with Adversarial Policy

Pytorch based library containing reinforcement learning agents with forward models and intrinsic motivation modules

Official PyTorch Implementation of Our Paper Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks

Playroom Framework: Reinforcement Learning Platform with Intrinsic Motivation

[IJCAI 2024] Constrained Intrinsic Motivation for Reinforcement Learning

Self-supervised Network Distillation is class of intrinsic motivation algorithms for reinforcement learning

An exploration of the effects of Intrinsic Motivation methods on RL algorithms using Atari games.

This is the code repository for my thesis