https://www.kyoto-u.ac.jp/ja/access/campus/yoshida/map6r_i.html

2018年1月11日に、オランダ・ラドバウド大学のMarcel van Gerven教授をお迎えして講演会を開催します。van Gerven先生は、計算神経科学や人工知能の手法を脳画像解析に応用する研究を活発に進めている気鋭の研究者です。本講演では、深層ニューラルネットワークを活用したヒトの認知研究のアプローチについてお話いただきます。

日時：2018年1月11日（木）16:30-18:00 場所：京都大学吉田キャンパス 医学部構内 先端科学研究棟1階 大セミナー室 https://www.kyoto-u.ac.jp/ja/access/campus/yoshida/map6r_i.html 参加費や登録は不要です。 本講演会は、科学研究費補助金基盤研究（S）「心的イメージの神経基盤の解明」（研究代表者：神谷之康、JP15H05710） の助成を受けて行われます

Neural networks as a window on human cognition

Marcel van Gerven Donders Institute for Brain, Cognition and Behaviour Department of Artificial Intelligence, Radboud University

In past years, modern neural networks have led to major breakthroughs in various areas of research. In the Artificial Cognitive Systems lab we have embraced the use of neural networks to further our understanding of human cognition and endowing machines with cognitive abilities. In this talk, I will touch upon the different ways in which neural networks can be used as a tool in neuroscience to model neural information processing. Particularly, I will describe our use of neural networks to construct encoding models in various domains. Next, I will describe how adversarial training of neural networks can be used to generate state of the art reconstructions of internal representations from brain activity patterns. I will also touch upon our recent work which uses neural networks to restore function in the blind and create empathic robots that are able to recognize someone’s personality. I will conclude the presentation with a general discussion of how to model natural intelligence with neural networks.