Dreamsmooth: Improving Model-based Reinforcement Learning Via Reward Smoothing

Abstract

Model-based reinforcement learning (MBRL) has gained much attention for its ability to learn complex behaviors in a sample-efficient way: planning actions by generating imaginary trajectories with predicted rewards. Despite its success, we found that surprisingly, reward prediction is often a bottleneck of MBRL, especially for sparse rewards that are challenging (or even ambiguous) to predict. Motivated by the intuition that humans can learn from rough reward estimates, we propose a simple yet effective reward smoothing approach, DreamSmooth, which learns to predict a temporally-smoothed reward, instead of the exact reward at the given timestep. We empirically show that DreamSmooth achieves state-of-the-art performance on long-horizon sparse-reward tasks both in sample efficiency and final performance without losing performance on common benchmarks, such as Deepmind Control Suite and Atari benchmarks.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylee2023dreamsmooth

Related papers

• Harmonydream: Task Harmonization Inside World Models (2023)3.46
• Breaking The Barrier: Enhanced Utility And Robustness In Smoothed DRL Agents (2024)0.00
• Learning The Reward Function For A Misspecified Model (2018)0.00
• Enhancing Control Policy Smoothness By Aligning Actions With Predictions From Preceding States (2026)0.00
• Planning With Exploration: Addressing Dynamics Bottleneck In Model-based Reinforcement Learning (2020)0.00
• World Model Agents With Change-based Intrinsic Motivation (2025)0.00
• DREAM: Deep Regret Minimization With Advantage Baselines And Model-free Learning (2020)0.00
• Multi-objective Model-based Policy Search For Data-efficient Learning With Sparse Rewards (2018)0.00