Výsledky vyhledávání - "Takahashi Ryota"

Akademický článek

Phosphine Susceptibility of Adult Megaselia scalaris (Loew) (Diptera: Phoridae)

Autor: Iwamoto Hirohide, Takahashi Ryota, Imai Toshihiro

Publikováno v: Contributions to Tobacco and Nicotine Research, Vol 31, Iss 2, Pp 101-105 (2022)

Megaselia scalaris (Loew) (Diptera: Phoridae) is a synanthropic fly that lives in a wide range of warmer regions globally. Although it is listed as a quarantine pest in the Eurasian Economic Union, phytosanitary protocols have not been established. T

Externí odkaz: https://doaj.org/article/dde44623bc5745b3bf31529b3f5f8248

Zobrazit plný text záznamu

Report

Deep RL with Hierarchical Action Exploration for Dialogue Generation

Autor: Cho, Itsugun, Takahashi, Ryota, Yanase, Yusaku, Saito, Hiroaki

Traditionally, approximate dynamic programming is employed in dialogue generation with greedy policy improvement through action sampling, as the natural language action space is vast. However, this practice is inefficient for reinforcement learning (

Externí odkaz: http://arxiv.org/abs/2303.13465

Zobrazit plný text záznamu

Akademický článek

Surface Chemistry Involved in Epitaxy of Graphene on 3C-SiC(111)/Si(111)

Autor: Abe Shunsuke, Handa Hiroyuki, Takahashi Ryota, Imaizumi Kei, Fukidome Hirokazu, Suemitsu Maki

Publikováno v: Nanoscale Research Letters, Vol 5, Iss 12, Pp 1888-1891 (2010)

Abstract Surface chemistry involved in the epitaxy of graphene by sublimating Si atoms from the surface of epitaxial 3C-SiC(111) thin films on Si(111) has been studied. The change in the surface composition during graphene epitaxy is monitored by in

Externí odkaz: https://doaj.org/article/8dc7c545703b45a4911e22e2151fcb0c

Zobrazit plný text záznamu

Report

A Personalized Dialogue Generator with Implicit User Persona Detection

Autor: Cho, Itsugun, Wang, Dongyang, Takahashi, Ryota, Saito, Hiroaki

Current works in the generation of personalized dialogue primarily contribute to the agent presenting a consistent personality and driving a more informative response. However, we found that the generated responses from most previous models tend to b

Externí odkaz: http://arxiv.org/abs/2204.07372

Zobrazit plný text záznamu