Convolutional fitted Q iteration for vision-based control problems
文献类型：会议论文
| 作者 | Zhao Dongbin ; Zhu Yuanheng; Lv Le; Chen Yaran; Zhang Qichao
|
| 出版日期 | 2016-11 |
| 会议日期 | 24-29 July 2016 |
| 会议地点 | Vancouver, BC, Canada |
| DOI | 10.1109/IJCNN.2016.7727794 |
| 英文摘要 | In this paper a deep reinforcement learning (DRL) method is proposed to solve the control problem which takes raw image pixels as input states. A convolutional neural network (CNN) is used to approximate Q functions, termed as Q-CNN. A pretrained network, which is the result of a classification challenge on a vast set of natural images, initializes the parameters of Q-CNN. Such initialization assigns Q-CNN with the features of image representation, so it is more concentrated on the control tasks. The weights are tuned under the scheme of fitted Q iteration (FQI), which is an offline reinforcement learning method with the stable convergence property. To demonstrate the performance, a modified Food-Poison problem is simulated. The agent determines its movements based on its forward view. In the end the algorithm successfully learns a satisfied policy which has better performance than the results of previous researches. |
| 源URL | [http://ir.ia.ac.cn/handle/173211/14476]  |
| 专题 | 复杂系统管理与控制国家重点实验室_深度强化学习 |
| 作者单位 | he State Key Laboratory of Management and Control for Complex Systems, In- stitution of Automation, Chinese Academy of Sciences, Beijing 100190, China. |
| 推荐引用方式 GB/T 7714 | Zhao Dongbin,Zhu Yuanheng,Lv Le,et al. Convolutional fitted Q iteration for vision-based control problems[C]. 见:. Vancouver, BC, Canada. 24-29 July 2016. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
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。