Exploration And Anti-exploration With Distributional Random Network Distillation

Abstract

Exploration remains a critical issue in deep reinforcement learning for an agent to attain high returns in unknown environments. Although the prevailing exploration Random Network Distillation (RND) algorithm has been demonstrated to be effective in numerous environments, it often needs more discriminative power in bonus allocation. This paper highlights the "bonus inconsistency" issue within RND, pinpointing its primary limitation. To address this issue, we introduce the Distributional RND (DRND), a derivative of the RND. DRND enhances the exploration process by distilling a distribution of random networks and implicitly incorporating pseudo counts to improve the precision of bonus allocation. This refinement encourages agents to engage in more extensive exploration. Our method effectively mitigates the inconsistency issue without introducing significant computational overhead. Both theoretical analysis and experimental results demonstrate the superiority of our approach over the orig

Stats

• citations0
• S2 citations—
• github stars17
• HF likes0
• heat score2.51
• arxiv keyyang2024exploration

Related papers

• Exploration By Random Distribution Distillation (2025)0.00
• Prend: Enhancing Intrinsic Motivation In Reinforcement Learning Through Pre-trained Network Distillation (2024)0.00
• Non-decreasing Quantile Function Network With Efficient Exploration For Distributional Reinforcement Learning (2021)4.52
• Random Latent Exploration For Deep Reinforcement Learning (2024)0.00
• Information-directed Exploration For Deep Reinforcement Learning (2018)0.00
• Pitfall Of Optimism: Distributional Reinforcement Learning By Randomizing Risk Criterion (2023)0.00
• Sampling Efficient Deep Reinforcement Learning Through Preference-guided Stochastic Exploration (2022)8.09
• Exploring More When It Needs In Deep Reinforcement Learning (2021)0.00