Discovery of First Principle Equations from Numeric Data

Hiroshi Motoda ,

Keynote speaker at AI'99
Institute of Scientific and Industrial Research, Osaka University (Japan)

**Special date** Monday 13 December at 11am

Abstract

We show that there is a method that ensures to derive the first principle from numeric data. Notion of scales and an interesting property that is deduced by dimensional analysis are the basis of the method. These two together with simple mathematics can constrain the form of admissible relations. The method works for phenomena for which nothing is known about the number of equations that are needed to describe them and no knowledge of dimensions of the variables involved is available. Many of the known first principles that involves several tens of variables and a few tens of equations have been derived by numerical experiments with noisy data.

Short resume

Hiroshi Motoda is a professor in the division of Intelligent Systems Science at the Institute of Scientific and Industrial Research of Osaka University since 1996. Before joining the university, he had been with Hitachi since 1967, participated as a senior researcher in research on core management, control and design of nuclear power reactors and expert systems for plant diagnosis at the Energy Research Laboratory and later reached the position of a senior chief research scientist at the Advanced Research Laboratory where he headed an AI group and conducted research on machine learning, knowledge acquisition and diagrammatic reasoning. In addition to these his current research interests include scientific knowledge discovery and data mining.

He received his Bs, Ms and PhD degrees in nuclear engineering from University of Tokyo. He was on the board of trustee of the Japan Society of Software Science and Technology (JSSST), the Japanese Society for Artificial Intelligence (JSAI) and the Japanese Cognitive Science Society (JCSS), the chair of SIG-KBS (Knowledge Based Systems) of JSAI, and on the editorial board of JSAI, JCSS and Knowledge Acquisition and IEEE Expert. He is now the chair of SIG-FAI (Fundmental AI) of JSAI and on the editorial board of Artificial Intelligence in Engineering, International Journal of Human-Computer Studies and Knowledge and Information Systems: An International Journal. He is a member of AAAI, IEEE Computer Society, JSAI, JSSST, IPSJ and JCSS.

To top

last updated December 01, 2003 10:37 AM
Sandrine.Balbo@cmis.csiro.au