Abstract:Preference-based Reinforcement Learning (PbRL) methods provide a solution to avoid reward engineering by learning reward models based on human preferences. However, poor feedback- and sample- efficiency still remain the problems that hinder the application of PbRL. In this paper, we present a novel efficient query selection and preference-guided exploration method, called SENIOR, which could select the meaningful and easy-to-comparison behavior segment pairs to improve human feedback-efficiency and accelerate policy learning with the designed preference-guided intrinsic rewards. Our key idea is twofold: (1) We designed a Motion-Distinction-based Selection scheme (MDS). It selects segment pairs with apparent motion and different directions through kernel density estimation of states, which is more task-related and easy for human preference labeling; (2) We proposed a novel preference-guided exploration method (PGE). It encourages the exploration towards the states with high preference and low visits and continuously guides the agent achieving the valuable samples. The synergy between the two mechanisms could significantly accelerate the progress of reward and policy learning. Our experiments show that SENIOR outperforms other five existing methods in both human feedback-efficiency and policy convergence speed on six complex robot manipulation tasks from simulation and four real-worlds. Videos can be found on our project website: this https URL
| Comments: | 8 pages, 8 figures, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025) |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2506.14648 [cs.RO] |
| (or arXiv:2506.14648v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2506.14648 arXiv-issued DOI via DataCite |
From: Hexian Ni [view email]
[v1]
Tue, 17 Jun 2025 15:42:19 UTC (5,129 KB)
[v2]
Thu, 21 May 2026 02:08:17 UTC (5,221 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。