Presentation | 2020-03-06 Considering a human mobility model inspired by reinforcement learning Yuutaro Iwai, Akihiro Fujihara, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Reinforcement learning is a machine learning framework that an agent repeatedly observes reward from environment as a result of action to choose actions with high reward expectations. On the other hand, mobility models that satisfy the statistical properties of human mobility patterns have been proposed. However, a method of generating a mobility model based on reinforcement learning has not been sufficiently studied. In this paper, a dataset containing mobility traces of taxi cabs in San Francisco is used to propose a method of generating a mobility model that learns action selection by giving rewards, such as fare and gasoline charge, and also the distribution of passengers' positions. The $epsilon$-greedy method was used as the reinforcement learning algorithm. The degree of learning was compared by changing the search parameter $epsilon$. As a result, it was found that the cumulative reward was highest when $epsilon=0$ because the taxi movement included an implicit search. The generalization of the mobility model generation method using reinforcement algorithm was also considered. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Human mobility models / Reinforcement learning / Data assimilation |
Paper # | IN2019-118 |
Date of Issue | 2020-02-27 (IN) |
Conference Information | |
Committee | NS / IN |
---|---|
Conference Date | 2020/3/5(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Royal Hotel Okinawa Zanpa-Misaki |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | General |
Chair | Yoshikatsu Okazaki(NTT) / Takuji Kishida(NTT-AT) |
Vice Chair | Akihiro Nakao(Univ. of Tokyo) / Kenji Ishida(Hiroshima City Univ.) |
Secretary | Akihiro Nakao(Osaka Pref Univ.) / Kenji Ishida(NTT) |
Assistant | Shinya Kawano(NTT) |
Paper Information | |
Registration To | Technical Committee on Network Systems / Technical Committee on Information Networks |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Considering a human mobility model inspired by reinforcement learning |
Sub Title (in English) | |
Keyword(1) | Human mobility models |
Keyword(2) | Reinforcement learning |
Keyword(3) | Data assimilation |
1st Author's Name | Yuutaro Iwai |
1st Author's Affiliation | Chiba Institute of Technology(CIT) |
2nd Author's Name | Akihiro Fujihara |
2nd Author's Affiliation | Chiba Institute of Technology(CIT) |
Date | 2020-03-06 |
Paper # | IN2019-118 |
Volume (vol) | vol.119 |
Number (no) | IN-461 |
Page | pp.pp.237-242(IN), |
#Pages | 6 |
Date of Issue | 2020-02-27 (IN) |