paper-with-me

홈 › Papers

ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization

2024-10-17 · Chen Bo Calvin Zhang, Zhang-Wei Hong, Aldo Pacchiano, Pulkit Agrawal

Reward shaping is critical in reinforcement learning (RL), particularly for complex tasks where sparse rewards can hinder learning. However, choosing effective shaping rewards from a set of reward functions in a computationally efficient manner remains an open challenge. We propose Online Reward Selection and Policy Optimization (ORSO), a novel approach that frames the selection of shaping reward function as an online model selection problem. ORSO automatically identifies performant shaping reward functions without human intervention with provable regret guarantees. We demonstrate ORSO's effectiveness across various continuous control tasks. Compared to prior approaches, ORSO significantly reduces the amount of data required to evaluate a shaping reward function, resulting in superior data efficiency and a significant reduction in computational time (up to 8 times). ORSO consistently identifies high-quality reward functions outperforming prior methods by more than 50% and on average identifies policies as performant as the ones learned using manually engineered reward functions by domain experts.

📄 PDF Abstract BibTeX arXiv:2410.13837

Code (1)

calvincbzhang/orso 공식 구현 pytorch

Tasks

continuous-controlContinuous ControlModel SelectionReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Chasing Stability: Humanoid Running via Control Lyapunov Function Guided Reinforcement Learning

2025-09-23 · Zachary Olkin, Kejun Li, William D. Compton, Aaron D. Ames arxiv

Achieving highly dynamic behaviors on humanoid robots, such as running, requires controllers that are both robust and precise, and hence difficult to design. Classical control methods offer valuable insight into how such…

Reinforcement Learning

KOI: Accelerating Online Imitation Learning via Hybrid Key-state Guidance

2024-08-06 · Jingxian Lu, Wenke Xia, Dong Wang, Zhigang Wang 외

Online Imitation Learning struggles with the gap between extensive online exploration space and limited expert trajectories, hindering efficient exploration due to inaccurate reward estimation. Inspired by the findings f…

Efficient ExplorationImitation LearningOptical Flow Estimation

Accelerating Exploration with Unlabeled Prior Data

2023-11-09 · NeurIPS 2023 11 · Qiyang Li, Jason Zhang, Dibya Ghosh, Amy Zhang 외

Learning to solve tasks from a sparse reward signal is a major challenge for standard reinforcement learning (RL) algorithms. However, in the real world, agents rarely need to solve sparse reward tasks entirely from scra…

Reinforcement Learning (RL)

MPC-Injection: Biasing Off-Policy Locomotion RL Toward Controller-Induced Behavior Basins

2026-06-24 · Roy Xing, Seyoung Ree, Brian Plancher arxiv

Reinforcement learning (RL) for locomotion frequently converges to locally optimal but undeployable behaviors, such as vibrating limbs or scooting on the torso, that maximize return without producing a usable gait. We pr…

Reinforcement Learning

Euphonium: Steering Video Flow Matching via Process Reward Gradient Guided Stochastic Dynamics

2026-02-04 · Ruizhe Zhong, Jiesong Lian, Xiaoyue Mi, Zixiang Zhou 외 arxiv

While online Reinforcement Learning has emerged as a crucial technique for aligning flow matching models with human preferences, current approaches are hindered by inefficient exploration during training rollouts. Relyin…

Text-to-Video GenerationReinforcement Learning