Totally model-free actor-critic recurrent neural-network reinforcement learning in non-Markovian domains
Crossref DOI link: https://doi.org/10.1007/s10479-016-2366-2
Published Online: 2016-11-08
Published Print: 2017-11
Update policy: https://doi.org/10.1007/springer_crossmark_policy
Mizutani, Eiji
Dreyfus, Stuart
Funding for this research was provided by:
Ministry of Science and Technology, Taiwan (104-2221-E-011-096)
License valid from 2016-11-08