Please wait a minute...
IET Cyber-Systems and Robotics
IET Cyber-Systems and Robotics  2021, Vol. 3 Issue (2): 150-163    DOI: 10.1049/csy2.12017
    
Gaussian processes non-linear inverse reinforcement learning
Gaussian processes non-linear inverse reinforcement learning
 全文: PDF 
摘要： The authors analyse a Bayesian framework for posing and solving inverse reinforcement learning (IRL) problems that arise in decision-making and optimisation settings. The authors propose a non-parametric Bayesian model using Gaussian process (GP) and preference graphs, which offer an effective and computationally efficient method for ill-posed IRL problems in large or infinite state space. This approach only requires a finite number of demonstrations that is much less than the amount required for approximating the feature expectation or value functions in previous IRL methods. The proposed learning framework is expressive as it relies on a Bayesian structure in which assumptions are explicit and changeable. It is also robust in that it formalises on convex optimisation, which retains the promise of computationally manageable implementations for practical problems. To deal with more realistic IRL problems where the dynamics is also unknown, the GP model can be easily combined with the method to learn the dynamics at the same time. Experimental results prove the superiority of the authors method to current prevailing IRL algorithms with fewer numbers of demonstrations in both discrete and continuous state space.
Abstract: The authors analyse a Bayesian framework for posing and solving inverse reinforcement learning (IRL) problems that arise in decision-making and optimisation settings. The authors propose a non-parametric Bayesian model using Gaussian process (GP) and preference graphs, which offer an effective and computationally efficient method for ill-posed IRL problems in large or infinite state space. This approach only requires a finite number of demonstrations that is much less than the amount required for approximating the feature expectation or value functions in previous IRL methods. The proposed learning framework is expressive as it relies on a Bayesian structure in which assumptions are explicit and changeable. It is also robust in that it formalises on convex optimisation, which retains the promise of computationally manageable implementations for practical problems. To deal with more realistic IRL problems where the dynamics is also unknown, the GP model can be easily combined with the method to learn the dynamics at the same time. Experimental results prove the superiority of the authors method to current prevailing IRL algorithms with fewer numbers of demonstrations in both discrete and continuous state space.
出版日期: 2021-05-04
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Qifeng Qiao
Xiaomin Lin

引用本文:

Qifeng Qiao, Xiaomin Lin. Gaussian processes non-linear inverse reinforcement learning. IET Cyber-Systems and Robotics, 2021, 3(2): 150-163.

链接本文:

https://www.zjujournals.com/iet-csr/CN/10.1049/csy2.12017    或    https://www.zjujournals.com/iet-csr/CN/Y2021/V3/I2/150

No related articles found!