Philip G.K. Reiser

Research Scientist



Research Interests

My principal research interests lie at the intersection of biology and and computer science, more specifically:

My PhD was concerned with the use of evolutionary algorithms to induce formulae in first order logic.


Guest lecturer: Introduction to Artificial Intelligence (415.365) (University of Auckland)
Tutor: Software Reliability and Design (415.232) (University of Auckland)
Tutor: Programming courses in C, Ada, Unix (University of Wales)
Tutor: Discrete mathematics (University of Wales)

Post-doctoral Training

The Robot Scientist

Funded by
Prof. Ross D. King (principal investigator, Aberystwyth)
Collaborators: Stephen Muggleton (Imperial), Steve Oliver (Manchester), Douglas Kell (UMIST)

We aim to provide a physical implementation of a scientific active learning system, and apply it to the problem of functional genomics. The system will: use machine learning to form an initial hypotheses set, devise experiments to select between competing hypotheses, direct a robot to physically perform these experiments, automatically analyse the experimental results, revise its hypothesis set in the light of the experimental results, and then repeat the cycle until the user's criteria for selection of the best hypothesis is met. In this way the system will automatically produce useful knowledge about genes of unknown function.

Towards Automated Preprocessing in Data-Mining

Funded by The Boeing Company
Dr. Patricia Riddle (principal investigator, Auckland)
Michael Goebel (PhD student, Auckland)
Collaborators: Mike Barley (Auckland)

Most machine learning researchers estimate that 80-90% of the work in a successful application of machine learning (or data-mining) is the preparation work (combining data from different sources, selecting and creating new attributes, choosing an appropriate representation for the data, and choosing an appropriate machine learning algorithm). Yet most machine learning research has focused on the 10-20% portion of the problem, running the machine learning algorithm. We plan to address this situation by investigating the data-engineering process. This will proceed with the development of data-engineering tools that will reduce the labour involved in implementing data-mining solutions.


(14) King, R.D. Whelan, K.E., Jones, F.M., Reiser, P.G.K., Bryant, C.H., Muggleton, S.H., Kell, D.B. and Oliver, S.G. (2004) A robot scientist: automated hypothesis generation and experimentation for functional genomics. Nature, 427, 247-252. [pdf, doi, Nature]

(13) Reiser, P.G.K., King, R.D. (2001) The Robot Scientist Platform. Technical report UWA-DCS-02-041. University of Wales. [ps, pdf]

(12) Bryant, C.H., Muggleton, S.H., Oliver, S.G., Kell, D.B., Reiser, P.G.K., King, R.D. (2001) Combining Inductive Logic Programming, Active Learning and Robotics to Discover the Function of Genes. Electronic Transactions in Artificial Intelligence 6(12). [ps, pdf]

(11) Reiser, P.G.K., King, R.D., Kell, D.B., Muggleton, S.H., Bryant, C.H., Oliver,S.G. (2001) Developing a Logical Model of Yeast Metabolism. Electronic Transactions in Artificial Intelligence 6(24). [ps, pdf]

(10) Reiser, P.G.K. and Riddle, P.J. (2001) Scaling up Inductive Logic Programming: an Evolutionary Wrapper Approach. Applied Intelligence. 15(3), 181-197. [ps, pdf, doi, Kluwer]

(9) Reiser, P.G.K. (1999) Evolutionary algorithms for learning formulae in first order logic. PhD Thesis. University of Wales, UK [ps, pdf].

(8) Reiser, P.G.K. and Freitas, A.A. (1999) A Survey of Evolutionary Algorithms for the Discovery of Classification Rules. (unpublished)

(7) Reiser, P.G.K. and Riddle, P.J. (1999) An Evolutionary Methodology for Exploratory Machine Learning In: Proceedings of the Genetic and Evolutionary Computation Conference. Florida. (submitted)

(6) Reiser, P.G.K. and Riddle, P.J. (1999) Evolution of Logic Programs: Part-of-Speech Tagging In: Proceedings of the Congress on Evolutionary Computation. Washington D.C. Vol 2. pp. 1338-1345. [ps, pdf, IEEE]

(5) Reiser, P.G.K. and Riddle, P.J. (1998) Evolving Logic Programs to Classify Chess-Endgame Positions In: Simulated Evolution and Learning. Canberra, Australia. Lecture Notes in Artificial Intelligence No. 1585. B. McKay et al (eds). Springer-Verlag. pp. 138-145. [ps, pdf, Springer]

(4) Reiser, P.G.K. (1998) Evolutionary Computation and the Tinkerer's Evolving Toolbox In: Proceedings of the First European Workshop on Genetic Programming. Paris, France. Lecture Notes in Computer Science. Wolfgang Banzhaf, Riccardo Poli, Marc Schoenauer and Terence C. Fogarty (eds). Springer-Verlag pp. 209-219. [ps, pdf, Springer]

(3) Reiser, P.G.K. (1998) Computational Models of Evolutionary Learning In: Apprentissage: des principes naturels aux methodes artificielles. Ritschard, Berchtold, Duc and Zighed (editors). Hermes, Paris. pp. 291-306. [ps, pdf, Lavoisier]

(2) Reiser, P.G.K. (1997) An Evolutionary Learning System for Knowledge Discovery (extended abstract) In: Abstracts of the International Association of Interdisciplinary Research Conference on Learning. Geneva, Switzerland pp.234-237.

(1) Reiser, P.G.K. (1997) EVIL1: a Learning System to Evolve Logical Theories In: Proceedings of the International Workshop on Logic Programming and Multi-Agents. Leuven, Belgium pp.95-101.

Research collaboration

Prof. Ted Baker, University of Auckland.
Dr. Guillem Bernat, University of York.
Prof. John Fraser, University of Auckland.
Dr. Alex Freitas, University of Kent.
Prof. Doug Kell, UMIST.
Prof. Ross D. King, University of Wales.
Prof. Stephen Muggleton, Imperial College, London.
Prof. Steve Oliver, University of Manchester.
Dr. Pat Riddle, University of Auckland, NZ.

Philip Reiser