Zobrazeno 1 - 10
of 45
pro vyhledávání: '"Yoshihiro Ichikawa"'
Publikováno v:
Handbook of Research on New Investigations in Artificial Life, AI, and Machine Learning ISBN: 9781799886860
The mission of this chapter is to formalize multi-objective reinforcement learning (MORL) problems where there are multiple conflicting objectives with unknown weights. The objective is to collect all Pareto optimal policies in order to adapt them fo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d51bb30711c018566ca04462ad84c7db
https://doi.org/10.4018/978-1-7998-8686-0.ch012
https://doi.org/10.4018/978-1-7998-8686-0.ch012
Publikováno v:
Human Interface and the Management of Information. Information in Intelligent Systems ISBN: 9783030226480
HCI (5)
HCI (5)
This paper describes solving multi-objective reinforcement learning problems where there are multiple conflicting objectives with unknown weights. Reinforcement learning (RL) is a popular algorithm for automatically solving sequential decision proble
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1b313fb1bc766da6a7803b32f937d2bc
https://doi.org/10.1007/978-3-030-22649-7_25
https://doi.org/10.1007/978-3-030-22649-7_25
Autor:
Yoshihiro Ichikawa, Yusuke Tajima, Hiroyasu Matsushima, Kiyohiko Hattori, Hiroyuki Sato, Keiki Takadama, Masaya Nakata
Publikováno v:
New Mathematics and Natural Computation. 11:201-215
This paper proposes the evolutionary algorithm (EA) for the uncertain evaluation function in which fitness values change even with the same input. In detail, the proposed method employs the probability model to acquire the appropriate attributes that
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a555a89cc5bcb38db1c2257fb2a7e332
https://doi.org/10.1142/s1793005715400062
https://doi.org/10.1142/s1793005715400062
Autor:
Yoshihiro Ichikawa, Keiki Takadama
Publikováno v:
Journal of Advanced Computational Intelligence and Intelligent Informatics. 17:926-931
This paper proposes the reinforcement learning agent that estimates internal rewards using external rewards in order to avoid conflict in multi-step dilemma problem. Intensive simulation results have revealed that the agent succeeds in avoiding local
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6142fe5c226a0aa43ea9bfe5036395e8
https://doi.org/10.20965/jaciii.2013.p0926
https://doi.org/10.20965/jaciii.2013.p0926
Autor:
Masayuki Otani, Keiji Sato, Keiki Takadama, Hiroyasu Matsushima, Kiyohiko Hattori, Yoshihiro Ichikawa
Publikováno v:
Transactions of the Society of Instrument and Control Engineers. 48:745-753
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::dbbdf49ab85d0d100fd6d373f459096c
https://doi.org/10.9746/sicetr.48.745
https://doi.org/10.9746/sicetr.48.745
Autor:
Yoshihiro Ichikawa, Keiki Takadama
Publikováno v:
Transactions of the Society of Instrument and Control Engineers. 48:764-772
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b5a45993ae819ac79deea85c86772da8
https://doi.org/10.9746/sicetr.48.764
https://doi.org/10.9746/sicetr.48.764
Autor:
Masayuki Otani, Keiki Takadama, Yoshihiro Ichikawa, Tomohiro Harada, Kiyohiko Hattori, Hiroyuki Sato
Publikováno v:
Journal of Advanced Computational Intelligence and Intelligent Informatics. 15:1175-1185
This paper focuses on an on-board computer (OBC) that evolves computer programs through bit inversion and targets analyzing robustness against bit inversion in registers. We also propose a new method that can change the number of computer programs dy
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::381b1edbbd884661eaa82c240cce43e3
https://doi.org/10.20965/jaciii.2011.p1175
https://doi.org/10.20965/jaciii.2011.p1175
Autor:
Isao Yuri, Sazo Tsuruzono, Tetsuo Tatsumi, Tohru Hisamatsu, Takero Fukudome, Yoshihiro Ichikawa
Publikováno v:
Key Engineering Materials. :481-486
An 8000 kW class Hybrid Gas Turbine (HGT) project, administered by the New Energy and Industrial Technology Development Organization (NEDO), was completed in March 2004. The targets of this project were improvement in thermal efficiency and output po
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d84afbebc2c37652b660a9bffe6072b1
https://doi.org/10.4028/www.scientific.net/kem.317-318.481
https://doi.org/10.4028/www.scientific.net/kem.317-318.481
Publikováno v:
Wireless Personal Communications. 35:353-364
A tapped delayed line adaptive array antenna (TDL-AAA) and a space-temporal simultaneous processing equalizer (ST-SPE) are proposed as simple space-temporal equalizers based on minimum mean square error (MMSE) criterion. The ST-SPE has a compact hard
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::781d24960813e4203534a1a5db27c541
https://doi.org/10.1007/s11277-005-7534-7
https://doi.org/10.1007/s11277-005-7534-7
Autor:
Tetsuo Tatsumi, Isao Yuri, Takero Fukudome, Yoshihiro Ichikawa, Sazo Tsuruzono, Tohru Hisamatsu
Publikováno v:
Key Engineering Materials. 287:10-15
Silicon nitride is one of the most practical candidates for ceramic gas turbines. The SN282 is silicon nitride material developed by Kyocera for gas turbines. Several new technologies have been developed to achieve materialization of ceramic gas turb
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3398a9f2bfcb03395286ad057a1b6207
https://doi.org/10.4028/www.scientific.net/kem.287.10
https://doi.org/10.4028/www.scientific.net/kem.287.10