Art

J-GLOBAL ID：202002239824690621   Reference number：20A1071466

A simple pseudo sample generation algorithm using multilayered neural networks-Continual learning with relearning of pseudo samples-

多層神経回路からの疑似記憶サンプル生成法の検討-疑似記憶の再学習による追加学習-

Author (2)： SATO KAZUYA (中部大) ,  YAMAUCHI KOICHIRO (中部大)
Material： 電子情報通信学会技術研究報告  (IEICE Technical Report (Institute of Electronics, Information and Communication Engineers))
Volume： 119  Issue： 453(NC2019 75-116)(Web)  Page： 169-174 (WEB ONLY)  Publication year： Feb. 26, 2020 
JST Material Number： S0532B  ISSN： 0913-5685  Document type： Proceedings
Article type： 原著論文  Country of issue： Japan (JPN)  Language： JAPANESE (JA)

Thesaurus term： *neural network(biomodel) ,  *forgetting ,  *machine learning ,  model ,  neural pathway ,  recall ,  learning ,  algorithm ,  discriminant function ,  experiment ,  memory(psychology)
Semi thesaurus term： deep neural network ,  neuron model ,  relearning ,  Neural Circuit ,  neural circuit model ,  incremental learning ,  evaluation experiment

JST classification (1)： Artificial intelligence  (JE08000Z)

Reference (10)：

• German I. Parisi, Ronald Kemker, Jose L. Part, Christopher Kanan, and Stefan Wermter. Continual lifelong learning with neural networks: A review. Neural Networks, 113:51-71, 2019.
• Timothee Lesort, Alexander Gepperth, Andrei Stoian, and David Filliat. Marginal replay vs conditional replay for continual learning. In Igor V. Tetko, Věra Kůrková, Pavel Karpov, and Fabian Theis, editors, Artificial Neural Networks and Machine Learning - ICANN2019, volume LNCS 11728, pages 466-480. Elsevier Science Publishers B. V. (North-Holland), September 2019.
• James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A. Rusu, Kieran Milan, John Quan, and Tiago Ramalho. Overcoming catastrophic forgetting in neural networks. Proceeding of the National Acacemy of United States of America, pages 1091-6490, March 2017.
• Sang-Woo Lee, Jin-Hwa Kim, Jaehyun Jun, Jung-Woo Ha, and Byoung-Tak Zhang. Overcoming catastrophic forgetting byincremental moment matching. In 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA, 2017.
• Alexander Gepperth and Florian Wiech. Simplified computation and interpretation of fisher matrices in incremental learning with deep neural networks. In Igor V. Tetko, Věra Kůrková, Pavel Karpov, and Fabian Theis, editors, Artificial Neural Networks and Machine Learning - ICANN2019, volume LNCS 11728, pages 481-494. Elsevier Science Publishers B. V. (North-Holland), September 2019.

Terms in the title (5)： 多層 ,  神経回路 ,  サンプル ,  学習 ,  追加学習