Proceedings of the KIEE Conference (대한전기학회:학술대회논문집)
- 1998.11b
- /
- Pages.407-409
- /
- 1998
A study of Temperal Difference Learning using Nonlinear Function Approximation
비선형 함수 근사화를 사용한 TD학습에 관한 연구
- Kwon, Jae-Cheol (Dept. of Electrical Eng. K.N.U.) ;
- Lee, Young-Seog (Young-jin Junior college) ;
- Kim, Dong-Ok (Dept. of Electrical Eng. K.N.U.) ;
- Seo, Bo-Hyeok (School of Electronic & Electrical Eng. K.N.U)
- Published : 1998.11.28
Abstract
This paper deals with temporal-difference learning that is a method for approximating long-term future cost as a function of current state in knowlege-poor environment, a function approximator is used to approximate the mapping from state to future cost, a linear function approximator is limited because mapping from state to future cost has a nonlinear characteristic, so a nonlinear function approximator is used to approximate the mapping from state to future cost in this paper, and that TD learning using a nonlinear function approximator is stable is proved.
Keywords