[Back to Results | New Search]

URN	etd-0727115-130716
Author	Yi-Chia Tseng
Author's Email Address	No Public.
Statistics	This thesis had been viewed 5371 times. Download 842 times.
Department	Electrical Engineering
Year	2015
Semester	1
Degree	Master
Type of Document
Language	English
Title	An Unified Approach to Inverse Reinforcement Learning by Oppositive Demonstrations
Date of Defense	2015-08-25
Page Count	45
Keyword	Apprenticeship Learning Feature weight Inverse Reinforcement learning Reward function Reinforcement learning
Abstract	Reinforcement learning (RL) techniques use a reward function to correct a learning agent to solve sequential decision making problems through interactions with a dynamic environment, but it is hard to design the reward function in complex problems. Its design difficulties promote the inverse reinforcement learning (IRL) by deriving from an expert’s demonstrations. It is assumed that the demonstrations are meaningful and reproducible. In this thesis, demonstrations of failure are not entirely useless. An unified method of combining oppositive demonstrations is proposed to teach the robot by showing inappropriate demonstrations or trying to exhibit unrelated behaviors, so as to the agent can deliberately avoid such bad situations and speed up the learning. According to the result of simulations, it is obvious that the performance of algorithm combined with demonstrations of failure is better than that has only good demonstrations. It is not only convenient to operate but also save a lot of learning time.
Advisory Committee	Tzuu-Hseng S. Li - chair Ming-Yi Ju - co-chair Yu-Jen Chen - co-chair Kao-Shing Huang - advisor
Files	etd-0727115-130716.pdf indicate access worldwide
Date of Submission	2015-08-27

[Back to Results | New Search]