B.S. from Yokohama National University (2003), M.S. from Tokyo Institute of Technology (2005), Ph.D from Tokyo Institute of Technology (2008)
2024 - : Japan Advanced Institute of Science and Technology , Graduate School of Advanced Science and Technology , Professor
2021 - : Seikei University , Visiting researcher
2018 - : 理化学研究所 , 革新知能統合センター , 客員研究員
2017 - 2024 : Japan Advanced Institute of Science and Technology , Graduate School of Advanced Science and Technology , Associate Professor
2014 - : Visting fuculty at IDIAP research institute
2011 - : Assistant Professor at Tokyo Institute of Technology
2008 - : Project Assistant Professor at Kyoto University
Human interfaces and interactions, Intelligent informatics
Data mining, Machine learning, Human dynamics, Multimodal Interaction, Social Signal Processing
Human dynamics and social signal processing based on multimodal machine learning and data mining, and it's application for communicative robot/ agen.
(1)Multimodal Interaction Modeling: Face to face conversation is a fundamental communication method for information sharing, decision making and consensus building. It has various kind of types: casual talking with friends, business meeting, negotiation, counseling, and so on. People send not only verbal information, but also nonverbal information each other. Their role in conversation, attitude (e.g. passive, active, cooperative), intention (e.g. agree), and emotional state sometimes can be observed from their multimodal (verbal and nonverbal) signals called social signals [A.Vinciarelli et al 2009]. Conversational states such as lively discussion and to be silent can be observed by fusing social signals of all members. My researches focus on building computational model of multimodal social signals (speech, gaze, gesture and so on.) by using speech signal processing, image processing, motion sensor processing and pattern recognition techniques. My research question is how these social signal patterns influence high level output and tacit knowledge such as output after consensus building, communication skills and explanation skills. These modeling techniques can be also used to develop a sensing module for conversational robots/agents.
(2) Human Dynamics Modeling: Recent progress in developing sensors: location sensors, for monitoring human motion and activity has become available for analyzing longitudinal human activity and it’s dynamics in real environment. A research focuses on analyzing of office worker’s activity from sensor environment set in office. Another research focuses on analyzing of driver’s behaviors (brake patterns, how to press the accelerator or the brakes) from sensor environment equipped in cars.
(3) Machine Learning and Data Mining: Phenomenon of multimodal interaction and human dynamics are observed as continuous multi-dimensional time-series data from multiple sensors. Machine learning techniques are important to build recognition model from these multidimensional time-series data set. It is difficult to define Social signal patterns and typical activity patterns in office and represent features of it. To discover the structure of these patterns, Data mining algorithm are also useful. In particular, we focus on developing time-series clustering, multidimensional motif discovery and change point detection algorithm and applied to find various patterns and structure of data
ACM, THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS, THE JAPANESE SOCIETY FOR ARTIFICIAL INTELLIGENCE, IEEE
ACM International Conference on Multimodal Interaction (ICMI 2016) , Organizing CommitteeLocal Organization ChairShogo Okada (Tokyo Institute of Technology, Japan) , 2016 - 2016 , 東京 お台場開催，マルチモーダルインタラクションのトップカンファレンス
・ Outstanding performance award（2位） , MIYAMA, 三山 有, 岡田 将吾 , 対話ロボットコンペティション2022 , 2022
・ Best Paper Runner-up Award , Yukiko Nakano, Eri Hirose, Tatsuya Sakato, Shogo Okada, Jean-Claude MARTIN , ACM International Conference on Multimodal Interaction (2022) , 2022
・ 物質・デバイス共同研究賞 , 岡田将吾, 駒谷和範 , 物質 デバイス領域共同研究拠点 , 2022