Title:
モデル生成装置、モデル生成方法、及び、プログラム
Document Type and Number:
Japanese Patent JP7251613
Kind Code:
B2
Abstract:
The model generation device generates model parameters corresponding to the model to be used and mediation parameter relevance information indicating the relevance between the model parameters of a plurality of source domains and the mediation parameters by using the learning data in the plurality of source domains. The model adjustment device generates target model parameters which correspond to the target domain and include the mediation parameters, based on the learned model parameters for each of the plurality of source domains and the mediation parameter relevance information. Then, the model adjustment device uses the evaluation data of the target domain to determine the mediation parameters included in the target model parameters.
Inventors:
Azusa Sawada
Takeshi Shibata
Takeshi Shibata
Application Number:
JP2021511083A
Publication Date:
April 04, 2023
Filing Date:
June 25, 2019
Export Citation:
Assignee:
NEC
International Classes:
G06N20/00; G06N3/08
Domestic Patent References:
| JP2018081404A |
Other References:
MORIYA, T., et al.,"Progressive Neural Network-based Knowledge Transfer in Acoustic Models",Proceedings of 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Con,2018年11月15日,Pages 998-1002,ISBN: 978-988-14768-5-2, .
ALUMAE, T.,"Multi-domain Neural Network Language Model",Proceedings of the 14th Annual Conference of the International Speech Communication Association (INT,[online],2013年08月29日,Pages 2182-2186, and and
ALUMAE, T.,"Multi-domain Neural Network Language Model",Proceedings of the 14th Annual Conference of the International Speech Communication Association (INT,[online],2013年08月29日,Pages 2182-2186,
Attorney, Agent or Firm:
Nakamura Satoshi
Tatsuo Egami
Hiroshi Miwa
Tatsuo Egami
Hiroshi Miwa