Don’t Always Learn From Scratch:Reinforcement Learning with Adaptability
碩士 === 國立臺灣大學 === 電機工程學研究所 === 107 === How to obtain good internal representations of high dimensional states is an important problem in learning subsequent reinforcement learning (RL) tasks. Some conventional methods tried to align the subsequent tasks by using resources other than the source domai...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2019
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/kgxyf6 |