第六届智能信息处理国际会议IIP2010
6th IFIP International Conference On
Intelligent Information Processing
13-16, October, 2010, Salford Greater Manchester, UK
Sponsored by
International Federation for Information Processing, IFIP TC12
Co-Sponsored by
University of Salford,United Kingdom
Institute of Computing Technology, Chinese Academy of Sciences
General Chairs
S. Vadera (UK) M. Musen(USA) R. Mizoguchi (Japan)
Program Chairs
Z. Shi (China) A. Aamodt (Norway) D. Leake (USA)
Keynotes Speakers
Case-Based Reasoning Tomorrow:
Provenance, the Web, and Cases in the Future of Intelligent Information Processing
David Leake
School of Informatics and Computing, Bloomington
Indiana University
Abstract: The World Wide Web and grid computing provide new opportunities and challenges for artificial intelligence. This talk examines how case-based reasoning can respond to these challenges by leveraging large-scale information sources. It highlights opportunities for exploiting naturally arising cases and augmenting them with additional open sources, to enable robust support for human reasoning. It illustrates with examples from current research, focusing especially on how CBR can leverage frameworks being developed in burgeoning research activity in provenance capture and storage.
Knowledge Mining Biological Network Models
S.H. Muggleton
Department of Computing
Imperial College London
Abstract: In this talk we survey work being conducted at the Centre for Integrative Systems Biology at Imperial College on the use of machine learning to build models of biochemical pathways. Within the area of Systems Biology these models provide graph-based descriptions of bio-molecular interactions which describe cellular activities such as gene regulation, metabolism and transcription. One of the key advantages of the approach taken, Inductive Logic Programming, is the availability of background knowledge on existing known biochemical networks from publicly available resources such as KEGG and Biocyc. The topic has clear societal impact owing to its application in Biology and Medicine. Moreover, object descriptions in this domain have an inherently relational structure in the form of spatial and temporal interactions of the molecules involved. The relationships include biochemical reactions in which one set of metabolites is transformed to another mediated by the involvement of an enzyme. Existing genomic information is very incomplete concerning the functions and even the existence of genes and metabolites, leading to the necessity of techniques such as logical abduction to introduce novel functions and invent new objects. Moreover, the development of active learning algorithms has allowed automatic suggestion of new experiments to test novel hypotheses. The approach thus provides support for the overall scientific cycle of hypothesis generation and experimental testing.
Multivariate Bandits and their Applications
John Shawe-Taylor
Centre for Computational Statistics and Machine Learning
University College, London,UK
Abstract: We will review the multi-armed bandit problem and its application to optimising click-through for Web site banners. We will present multi-variate extensions to the basic bandit technology including the use of Gaussian Processes to model relations between different arms. This leads to the consideration of infinitely many arms as well as applications to grammar learning and optimisation.
附件下载: