paper-with-me

홈 › Papers

Intrinsic Reward Policy Optimization for Sparse-Reward Environments

2026-01-29 · Minjae Cho, Huy Trong Tran arxiv

Exploration is essential in reinforcement learning as an agent relies on trial and error to learn an optimal policy. However, when rewards are sparse, naive exploration strategies, like noise injection, are often insufficient. Intrinsic rewards can also provide principled guidance for exploration by, for example, combining them with extrinsic rewards to optimize a policy or using them to train subpolicies for hierarchical learning. However, the former approach suffers from unstable credit assignment, while the latter exhibits sample inefficiency and sub-optimality. We propose a policy optimization framework that leverages multiple intrinsic rewards to directly optimize a policy for an extrinsic reward without pretraining subpolicies. Our algorithm -- intrinsic reward policy optimization (IRPO) -- achieves this by using a surrogate policy gradient that provides a more informative learning signal than the true gradient in sparse-reward environments. We demonstrate that IRPO improves performance and sample efficiency relative to baselines in discrete and continuous environments, and formally analyze the optimization problem solved by IRPO. Our code is available at https://github.com/Mgineer117/IRPO.

📄 PDF Abstract BibTeX arXiv:2601.21391

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Model Ensemble-Based Intrinsic Reward for Sparse Reward Reinforcement Learning

2019-09-25 · Giseung Park, Whiyoung Jung, Sungho Choi, Youngchul Sung

In this paper, a new intrinsic reward generation method for sparse-reward reinforcement learning is proposed based on an ensemble of dynamics models. In the proposed method, the mixture of multiple dynamics models is use…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Count Counts: Motivating Exploration in LLM Reasoning with Count-based Intrinsic Rewards

2025-10-18 · Xuan Zhang, Ruixiao Li, Zhijian Zhou, Long Li 외 arxiv

Reinforcement Learning (RL) has become a compelling way to strengthen the multi step reasoning ability of Large Language Models (LLMs). However, prevalent RL paradigms still lean on sparse outcome-based rewards and limit…

Reinforcement Learning

Action-Dependent Optimality-Preserving Reward Shaping

2025-05-19 · Grant C. Forbes, JianXun Wang, Leonardo Villalobos-Arias, Arnav Jhala 외

Recent RL research has utilized reward shaping--particularly complex shaping rewards such as intrinsic motivation (IM)--to encourage agent exploration in sparse-reward environments. While often effective, ``reward hackin…

Montezuma's Revenge

SmoothVLA: Aligning Vision-Language-Action Models with Physical Constraints via Intrinsic Smoothness Optimization

2026-03-14 · Jiashun Li, Xiaoyu Shi, Hong Xie, Mingsheng Shang 외 arxiv

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for robotic manipulation. However, existing post-training methods face a dilemma between stability and exploration: Supervised Fine-Tuning (SFT) is …

Reinforcement Learning

Modeling Affect-based Intrinsic Rewards for Exploration and Learning

2019-12-01 · Dean Zadok, Daniel McDuff, Ashish Kapoor

Positive affect has been linked to increased interest, curiosity and satisfaction in human learning. In reinforcement learning, extrinsic rewards are often sparse and difficult to define, intrinsically motivated learning…

Reinforcement Learning