Gan-based Intrinsic Exploration For Sample Efficient Reinforcement Learning

Abstract

In this study, we address the problem of efficient exploration in reinforcement learning. Most common exploration approaches depend on random action selection, however these approaches do not work well in environments with sparse or no rewards. We propose Generative Adversarial Network-based Intrinsic Reward Module that learns the distribution of the observed states and sends an intrinsic reward that is computed as high for states that are out of distribution, in order to lead agent to unexplored states. We evaluate our approach in Super Mario Bros for a no reward setting and in Montezuma's Revenge for a sparse reward setting and show that our approach is indeed capable of exploring efficiently. We discuss a few weaknesses and conclude by discussing future works.

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keykamar2022gan

Related papers

• Generative Adversarial Exploration For Reinforcement Learning (2022)0.00
• Information Content Exploration (2023)0.00
• Go-explore: A New Approach For Hard-exploration Problems (2019)0.00
• Never Explore Repeatedly In Multi-agent Reinforcement Learning (2023)0.00
• Generative Adversarial Imagination For Sample Efficient Deep Reinforcement Learning (2019)0.00
• Learning Off-policy With Model-based Intrinsic Motivation For Active Online Exploration (2024)0.00
• Redeeming Intrinsic Rewards Via Constrained Optimization (2022)0.00
• R\'enyi State Entropy For Exploration Acceleration In Reinforcement Learning (2022)0.00