A meta-learning method based on temporal difference error

Lecture notes in computer science Volume 5863 Page 530-537

published_at 2009

Title

Creators Kobayashi Kunikazu

Creators Mizoue Hiroyuki

Creators Kuremoto Takashi

Creators Obayashi Masanao

Creator Keywords

reinforcement learning meta-parameter meta-learning TD-error maze search problem inverted pendulum control problem

Languages eng

Resource Type journal article

Publishers Springer

Date Issued 2009

File Version Not Applicable (or Unknown)

Access Rights metadata only access

Relations

[ISSN]0302-9743

[NCID]AA0071599X

info:doi/10.1007/978-3-642-10677-4_60

[isVersionOf] [URI]http://link.springer.com/chapter/10.1007/978-3-642-10677-4_60

[isVersionOf] [URI]http://www.springerlink.com/content/0302-9743/

Schools 大学院理工学研究科（工学）