Takashi Washio: Profile

First name: Takashi
Family name: Washio
Sex: Male
Personal Hobby: Sking

Office address:
The Institute of Scientific and Industrial Research
Osaka University
8-1 Mihogaoka, Ibaraki, Osaka, 567 Japan
Telephone: +81-6-6879-8540
Fax: +81-6-6879-8544
Email: washio(at)ar.sanken.osaka-u.ac.jp

Qualification: Doctor of Engineering
Position: Professor

Research interests:

• Research on Machine Learning and Data Mining Principles
• Measurement-oriented Machine Learning and Advanced Sensing
• Synthesis of Machine Learning and Simulation
• Data Mining and Knowledge Discovery from in massive data
• Discovery of regularity and concept embeded in massive data
• Discovery of causal information from data
• Industrial and scientific applications of machine learning and datamining

My Recent Academic Publications on Basic AI Research

Participating Academic Society:

• American Association for Artificial Intelligence (AAAI)
• IEEE Computer Society(IEEE CS)
• Society for Industrial and Applied Mathematics (SIAM)
• Japanese Society for Artificial Intelligence(JSAI)
• Japan Statistical Society(JSS)
• The Institute of Electronics, Information and Communication Engineers (IEICE)
• Information Processing Society of Japan(IPSJ)
• The Society of Instrument and Control Engineers (SICE)

Awards

• January, 2016 IBM Faculty Award
  
• June, 2009 Japanese Society for Artificial Intelligence (JSAI) Contribution Award
  
• March, 2002 JCAC(Journal of Computer Aided Chemistry) Journal Award (2002)
  
  • Applying the Apriori-based Graph Mining Method to Mutagenesis Data Analysis,
    Journal of Computer Aided Chemistry, Vol.2, pp.87-92, 2001
• May, 2001 Japanese Society for Artificial Intelligence (JSAI) Best Paper Awards
  
  • Scientific Law Discovery based on Scale Type Constraint
• November, 1999 Best Paper Awards for the Annual Conference of Japanese Society for Artificial Intelligence
  
  • Scientific Law Discovery from Observed Data and its Application to Socio-psychology
• October, 1998 Best Paper Awards for Annual Conference of Japanese Society for Artificial Intelligence
  
  • The Extended SDS: A Model of Scientific Discovery for Simultaneous Equation Systems
• June, 1997 Best Paper Awards for Annual Conference of Japanese Society for Artificial Intelligence
  
  • A Data-driven Method to Discover the First Principle of Complex Systems
• March, 1996 The Best Paper Award of Atomic Energy Society of Japan
  
  • Criteria of Diversity Evaluation for Intelligent Diagnosis of Nuclear Power Plants
• August, 1995 Best Paper Awards for Annual Conference of Japanese Society for Artificial Intelligence
  
  • A Method of History-Oriented Envisioning by Introducing Temporal Logic
• February, 1988 Premium Paper Awards of the Society of Instrument and Control Engineers
  
  • An Attempt of Fuzzy Qualitative Reasoning

Research Experience

• August, 2006 - Present
  Professor,
  Department of Advanced Reasoning, Division of Intelligent Systems Science,
  Institute for Scientific and Industrial Research (I.S.I.R.), Osaka University
  
  • Data Mining and Knowledge Discovery from in massive data
  • Discovery of regularity and concept embeded in massive data
  • Discovery of causal information from data
  • Industrial and scientific applications of Datamining and Knowledge Discovery


• June, 1996 - July, 2006
  Associate Professor,
  Department of Advanced Reasoning, Division of Intelligent Systems Science,
  Institute for Scientific and Industrial Research (I.S.I.R.), Osaka University
  
  • Discovery of a first principle equation from experimental and numerical data
  • Discovery of a first principle equation from observed and numerical data
  • Discovery of first principle simultaneous equations from experimental and numerical data
  • Discovery of first principle simultaneous equations from observed and numerical data
  • Complete search for extracting frequent patterns from graph structured data by Apriori-based graph mining (AGM)
  • Extention of Apriori-based graph mining (AGM) toward the application to wider areas
  • Discretization of numeric data by Akaike's information criterion (AIC) and minimum description length (MDL)
  • Reduction of association rules by maximal inference sets and Apriori algorithm
  • Feature generation method based on association rules
  • Fast classification method by using sub-sampling and Bagging
  • Identification and analysis of non-equilibrium thermodynamics from time series data analysis
  • Efficient approximate method of extracting typical patterns from graph structured data by graph based induction (GBI)
  • Integration of incremental knowledge acquisition from human expert and
    incremental machine learning from data in unstable environment within the framework of Ripple Down Rule method
  • Models of synthesis, its language and algorithm


• April, 1994 - May, 1996
  Senior Researcher, Safety Engineering Divison, MitsubishiResearch Institute Inc.
  
  • Diagnosis for multiple faults of components and sensor by using model based diagnosis
  • History-oriented envisioning to predict behaviors of process systems in qualitative reasoning
  • Abduction-based reasoning among multiple axiomatic systems
  • Worm-type agents for intelligent operation of large-scale man-machine systems
  • Organization learning in multi-agent systems
  • Autonomous recovery execution in nuclear power plant by agent
  • Diversity criteria for control and sensing instruments of nuclear power plants
  • Adaptive microphone array technique for remote monitoring of components in nuclear power plants


• February, 1990 - March, 1994
  Junior Researcher, Researcher, Safety Engineering Divison, Mitsubishi Research Institute Inc.
  
  • Causal reasoning based on inherent causal structure of physical law formulae
  
  
  • Human reliability analysis based on fuzzy integral modeling
  
  
  • Decision support for operability improvement and maintenance planning by analytic hierarchy process (AHP) methodology


• April, 1988 - January, 1990
  Visiting Researcher in Nuclear Reactor Laboratory, Massachusetts Institute of Technology (MIT)
  
  • Development of a quantitative and symbolic causal reasoning method for spacecraft nuclear reactors
  
  
  • Derivation of exogenously-driven causality based on assumptive structural equations
  • Stability considerations concerning the implementation of the MIT-SNL period-generated minimum time control laws

Academic Activities

• Editorial Board
  I served many editorial board members of prestigious international journals including
  Data Mining and Knowledge Discovery (DMKD), Knowledge and Information Systems: An International Journal (KAIS),
  Data Science and Engineering (DSE), New Generation Computing (NGC) and Journal of Atomic Energy Society of Japan (JAESJ)


• Steering Committee, Program Chairs and Organizers
  I served and am serving many steering committees, chairs and organizers of prestigious international conferences,
  workshops and contests including
  
  • Stearing committees of
    IEEE International Conference on Data Mining (ICDM) and
    Pacific Asia Knowledge Discovery and Data Mining (PAKDD),
  • Program chairs of
    Pacific Asia Knowledge Discovery and Data Mining 2008 (PAKDD2008),
    SIAM Conference on Data Mining 2011 (SDM11),
    Pacific Asia Knowledge Discovery and Data Mining 2016 (PAKDD2016),
  • Technical track chair of
    IEEE International Conference on Data Science and Advanced Analytics (IEEE DSAA2015).
  
  
• Program Committee
  I served and am serving many program committees of prestigious international conferences and workshops
  including SIG-KDD, ICDM, ECMLPKDD, SDM, PAKDD, ICML, NIPS, IJCAI, AAAI, DS and so on.