Knowledge Commons of Institute of Automation,CAS
Addressing Reward Engineering for Deep Reinforcement Learning on Multi-stage Task | |
Chen, Bin1,2; Su, Jianhua1 | |
2019-12 | |
会议名称 | International Conference on Neural Information Processing |
会议日期 | 2019-12 |
会议地点 | Australia |
摘要 | In the field of robotics, it is a challenge to deal with multistage tasks based on Deep reinforcement learning (Deep RL). Previous researches have shown manually shaping a reward function could easily result in sub-optimal performance, hence choosing a sparse reward is a natural and sensible decision in many cases. However, it is rare for the agent to explore a non-zero reward with the increase of the horizon under the sparse reward, which makes it difficult to learn an agent to deal with multi-stage task. In this paper, we aim to develop a Deep RL based policy through fully utilizing the demonstrations to address this problem. We use the learned policy to solve some difficult multi-stage tasks, such as picking-and-place, stacking blocks, and achieve good results. A video of our experiments can be found at: https://youtu.be/6BulNjqDg3I. |
收录类别 | EI |
七大方向——子方向分类 | 智能机器人 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/39041 |
专题 | 多模态人工智能系统全国重点实验室_机器人理论与应用 |
通讯作者 | Su, Jianhua |
作者单位 | 1.The State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences 2.The school of Artificial Intelligence, University of Chinese Academy of Sciences |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Chen, Bin,Su, Jianhua. Addressing Reward Engineering for Deep Reinforcement Learning on Multi-stage Task[C],2019. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Chen-Su2019_Chapter_(1169KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[Chen, Bin]的文章 |
[Su, Jianhua]的文章 |
百度学术 |
百度学术中相似的文章 |
[Chen, Bin]的文章 |
[Su, Jianhua]的文章 |
必应学术 |
必应学术中相似的文章 |
[Chen, Bin]的文章 |
[Su, Jianhua]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论