LEVERAGE EXPERIMENTS FROM OTHER TASKS TO SPEED UP TRAINING
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 107 === Although the deep reinforcement learning (RL) approach has achieved impressive results in a variety of video games, training by RL still requires a lot of time and computational resources since it is difficult to extract information from sparse reward by rando...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2018
|
Online Access: | http://ndltd.ncl.edu.tw/handle/bagb6j |