TITLE: Combining Knowledge Acquisition and Machine Learning
PRESENTER: JP Bekmann
AFFILIATION: Computer Science and Engineering, University of New South Wales,National ICT Australia.
DATE:Friday 12th December 2003
TIME: 12:00 noon - 1:00pm
PLACE: Seminar Room K17
The aim of this research is to attempt the integration of a machine learning and a knowledge acquisition process. The aim is to complement the strengths of both disciplines: leveraging a human's strategic and heuristic knowledge, and supporting it with powerful empirical evaluation and search in the machine learner. An approach is presented with which one can develop heuristic algorithms for NP-hard combinatorial problems. Specifically, a demonstration is made of finding heuristic layout algorithms for detailed channel routing in VLSI chips. The machine learning techniques are based on evolutionary algorithms, while knowledge acquisition is done according to the ripple down rule paradigm.
BIOGRAPHY OF SPEAKER:
JP Bekmann is a PhD student in Artificial Intelligence at the School of Computer Science and Engineering at UNSW, he is also part of the Intelligent Systems group at National ICT Australia. His area of research is in machine learning and knowledge acquisition. In a past life he has held various positions as software developer, including work on a Java virtual machine and compiler, cryptographic protocols and network drivers. He currently has no pets.
School of Computer Science & Engineering, UNSW.