Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
AI |
2023-09-12 14:55 |
Hokkaido |
|
Event-Driven Reinforcement Learning with Semi Markov Models for Stable Air-Conditioning Control Hayato Chujo, Arai Sachiyo (Chiba Univ) AI2023-16 |
This study deals with air conditioning control that optimizes room temperature by switching heaters on/off. The control ... [more] |
AI2023-16 pp.83-86 |
DC, SS |
2022-10-25 10:00 |
Fukushima |
(Primary: On-site, Secondary: Online) |
A note on performance and sensitivity analysis of self-adaptive systems using parametric Markov decision processes Junjun Zheng, Hiroyuki Nakagawa, Tatsuhiro Tsuchiya (Osaka Univ.) SS2022-21 DC2022-27 |
This paper considers the sensitivity analysis for a self-adaptive system with uncertain parameters. The system behavior ... [more] |
SS2022-21 DC2022-27 pp.1-5 |
IBISML |
2021-03-03 14:25 |
Online |
Online |
Markov Decision Processes for Simultaneous Control of Multiple Objects with Different State Transition Probabilities in Each Cluster Yuto Motomura, Akira Kamatsuka, Koki Kazama, Toshiyasu Matsushima (Waseda Univ.) IBISML2020-49 |
In this study, we propose an extended MDP model, which is a Markov decision process model with multiple control objects ... [more] |
IBISML2020-49 pp.47-54 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Approximate Value Iteration Algorithms for Partially Observable Markov Decision Processes in Geometric Dual Representation Hiroshi Tsukahara, Mitsuru Anbai, Makoto Oobayashi (Denso IT Lab.) IBISML2016-71 |
We propose new approximate algorithms for the value iteration of partially observable Markov decision
processes (POMDPs... [more] |
IBISML2016-71 pp.177-184 |
SIS |
2016-06-09 11:20 |
Hokkaido |
Kushiro Tourism and Convention cent. |
A Note on Teaching Strategies Using Markov Decision Processes Yasunari Maeda, Masakiyo Suzuki (KIT) SIS2016-2 |
In this research Markov decision processes(MDP) with unknown states are used in order to represent lectures. Effectivene... [more] |
SIS2016-2 pp.7-10 |
CAS, SIP, MSS, VLD, SIS [detail] |
2014-07-10 14:45 |
Hokkaido |
Hokkaido University |
[Tutorial Lecture]
Markov Decision Processes and Its Applications Yasunari Maeda, Masakiyo Suzuki (Kitami Inst. of Tech.) CAS2014-31 VLD2014-40 SIP2014-52 MSS2014-31 SIS2014-31 |
There are many research on Markov decision processes in the areas of operations research and artificial intelligence. Th... [more] |
CAS2014-31 VLD2014-40 SIP2014-52 MSS2014-31 SIS2014-31 pp.163-168 |
NC, NLP |
2013-01-24 10:10 |
Hokkaido |
Hokkaido University Centennial Memory Hall |
Analysis of Medical Treatment Data using Inverse Reinforcement Learning Hideki Asoh, Masanori Shiro, Toshihiro Kamishima, Shotaro Akaho (AIST), Takahide Kohro (Univ. of Tokyo Hospital) NLP2012-106 NC2012-96 |
It is an important issue to utilize large amount of medical records which are accumulated on medical information systems... [more] |
NLP2012-106 NC2012-96 pp.13-17 |
IBISML |
2012-11-07 15:30 |
Tokyo |
Bunkyo School Building, Tokyo Campus, Tsukuba Univ. |
Robustness of time-consistent Markov decision processes Takayuki Osogami (IBM Japan) IBISML2012-40 |
We show that an optimal policy for a Markov decision process (MDP) can be found with dynamic programming, when the objec... [more] |
IBISML2012-40 pp.45-52 |
IBISML |
2012-03-12 15:05 |
Tokyo |
The Institute of Statistical Mathematics |
Model Selection of Indirect Value Function Estimation Masahiro Kohjima (Tokyo Tech) IBISML2011-93 |
Reinforcement learning is a method to obtain a policy which maximizes expected return and is applied to wide range of re... [more] |
IBISML2011-93 pp.43-48 |
IBISML |
2011-11-09 15:45 |
Nara |
Nara Womens Univ. |
Active Value Function Estimation Based On Transition Probability Estimation Masahiro Kohjima (Tokyo Tech) IBISML2011-51 |
It is considered to be a great importance in reinforcement learning to estimate value function precisely. In this study,... [more] |
IBISML2011-51 pp.61-66 |
NC, IPSJ-BIO [detail] |
2011-06-24 16:30 |
Okinawa |
50th Anniversary Memorial Hall, University of the Ryukyus |
Solving POMDPs using Restricted Boltzmann Machines with Echo State Networks Makoto Otsuka, Junichiro Yoshimoto, Stefan Elfwing, Kenji Doya (OIST) NC2011-19 |
A partially observable Markov decision process (POMDP) can be solved in a model-based way using explicit knowledge of th... [more] |
NC2011-19 pp.143-148 |
SP |
2010-07-24 15:10 |
Miyagi |
Ryokusui-tei (Sendai) |
Spoken Dialogue Manager in Car Navigation System Using Partially Observable Markov Decision Processes with Hierarchical Reinforcement Learning Yasuhide Kishimoto, Tetsuya Takiguchi, Yasuo Ariki (Kobe Univ.) SP2010-43 |
In this paper,we propose a dialogue manager in a car navigation systems using Partially Observable Markov Decision Proce... [more] |
SP2010-43 pp.49-54 |
SIS |
2010-06-10 16:20 |
Hokkaido |
Abashiri Public Auditorium |
[Invited Talk]
Managing Credit Lines Using Markov Decision Processes Yasunari Maeda, Masakiyo Suzuki (Kitami Inst. of Tech.) SIS2010-13 |
In previous research Markov decision processes has been applied to managing credit lines. And an expected total discount... [more] |
SIS2010-13 pp.71-75 |
NC, MBE (Joint) |
2010-03-09 14:10 |
Tokyo |
Tamagawa University |
Introducing a New Function to Save the Trouble of Parameter Tuning of Softmax Action-Selection Kenji Ono, Kazunori Iwata, Akira Hayashi, Nobuo Suematsu (Hiroshima City Univ.) NC2009-106 |
Markov decision processes are one of the most popular frameworks for reinforcement learning. The entropy of probability ... [more] |
NC2009-106 pp.107-112 |
R |
2009-11-20 15:10 |
Osaka |
|
On an Optimal Maintenance Policy for Two-state POMDP Model with Multiple Observations Kenichi Hayashi, Nobuyuki Tamura, Tetsushi Yuge, Shigeru Yanagi (N.D.A) R2009-44 |
This study considers a system which can be in either a GOOD or BAD state
and which deteriorates stochastically in discr... [more] |
R2009-44 pp.17-22 |
NC |
2007-03-14 15:30 |
Tokyo |
Tamagawa University |
A Study of Policy-Gradient Methods in Non-Markov Decision Porcesses
-- Application to a Curling Game -- Harukazu Igarashi (Shibaura Inst. Tech.), Seiji Ishihara (Kinki Univ.), Masaomi Kimura (Shibaura Inst. Tech.) |
There are two approaches to reinforcement learning: value-based methods and policy-gradient methods. Baird and Moore pro... [more] |
NC2006-148 pp.179-184 |