Zobrazeno 1 - 10
of 83
pro vyhledávání: '"offline meta-reinforcement learning"'
Offline meta reinforcement learning (OMRL) has emerged as a promising approach for interaction avoidance and strong generalization performance by leveraging pre-collected data and meta-learning techniques. Previous context-based approaches predominan
Externí odkaz:
http://arxiv.org/abs/2405.12001
Autor:
Li, Lanqing, Zhang, Hai, Zhang, Xinyu, Zhu, Shatong, Yu, Yang, Zhao, Junqiao, Heng, Pheng-Ann
Publikováno v:
38th Conference on Neural Information Processing Systems (NeurIPS 2024)
As a marriage between offline RL and meta-RL, the advent of offline meta-reinforcement learning (OMRL) has shown great promise in enabling RL agents to multi-task and quickly adapt while acquiring knowledge safely. Among which, context-based OMRL (CO
Externí odkaz:
http://arxiv.org/abs/2402.02429
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Generalization and sample efficiency have been long-standing issues concerning reinforcement learning, and thus the field of Offline Meta-Reinforcement Learning~(OMRL) has gained increasing attention due to its potential of solving a wide range of pr
Externí odkaz:
http://arxiv.org/abs/2312.15909
Autor:
Gao, Yunkai, Zhang, Rui, Guo, Jiaming, Wu, Fan, Yi, Qi, Peng, Shaohui, Lan, Siming, Chen, Ruizhi, Du, Zidong, Hu, Xing, Guo, Qi, Li, Ling, Chen, Yunji
Offline meta-reinforcement learning (OMRL) utilizes pre-collected offline datasets to enhance the agent's generalization ability on unseen tasks. However, the context shift problem arises due to the distribution discrepancy between the contexts used
Externí odkaz:
http://arxiv.org/abs/2311.03695
Recent offline meta-reinforcement learning (meta-RL) methods typically utilize task-dependent behavior policies (e.g., training RL agents on each individual task) to collect a multi-task dataset. However, these methods always require extra informatio
Externí odkaz:
http://arxiv.org/abs/2305.19529
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
HAN Xu, WU Feng
Publikováno v:
Jisuanji kexue yu tansuo, Vol 17, Iss 8, Pp 1917-1927 (2023)
Traditional reinforcement learning algorithms require lots of online interaction with the environment for training and cannot effectively adapt to changes in the task environment, making them difficult to apply to real-world problems. Offline meta-re
Externí odkaz:
https://doaj.org/article/71335025c9734a6c82d2c2da77e27912
Autor:
Lin, Runji, Li, Ye, Feng, Xidong, Zhang, Zhaowei, Fung, Xian Hong Wu, Zhang, Haifeng, Wang, Jun, Du, Yali, Yang, Yaodong
The pretrain-finetuning paradigm in large-scale sequence models has made significant progress in natural language processing and computer vision tasks. However, such a paradigm is still hindered by several challenges in Reinforcement Learning (RL), i
Externí odkaz:
http://arxiv.org/abs/2211.08016
Autor:
Yuan, Haoqi, Lu, Zongqing
We study offline meta-reinforcement learning, a practical reinforcement learning paradigm that learns from offline data to adapt to new tasks. The distribution of offline data is determined jointly by the behavior policy and the task. Existing offlin
Externí odkaz:
http://arxiv.org/abs/2206.10442