no code implementations • 13 Feb 2022 • Yuto Omae, Makoto Sasaki, Jun Toyotani, Kazuyuki Hara, Hirotaka Takahashi
We used the solution to determine the target rate of the vaccination for decreasing the infection numbers of the COVID-19 Delta variant (B. 1. 617) in Japan.
no code implementations • 17 Oct 2020 • Yuto Omae, Jun Toyotani, Kazuyuki Hara, Yasuhiro Gon, Hirotaka Takahashi
As of 2020, COVID-19 is spreading in the world.
no code implementations • 30 Aug 2020 • Yuto Omae, Jun Toyotani, Kazuyuki Hara, Yasuhiro Gon, Hirotaka Takahashi
In this study, we describe the simulator and the effectiveness of the app in various scenarios.
Computers and Society Multiagent Systems I.2.11; I.6.3; I.6.5; I.6.6
no code implementations • 9 Nov 2017 • Kazuyuki Hara
This paper presents our investigation on the effect of dropout in online learning.
no code implementations • 20 Jun 2017 • Kazuyuki Hara, Daisuke Saitoh, Hayaru Shouno
We find that the process of combining the neglected hidden units with the learned network can be regarded as ensemble learning, so we analyze dropout learning from this point of view.
no code implementations • 20 Jun 2017 • Kazuyuki Hara, Kentaro Katahira, Masato Okada
The value of the objective function for an unperturbed output is called a baseline.