Peter Lane

Contact | Publications | Research | Software


UH Profile - my research is mainly in the area of learning algorithms and their applications:

  • Cognitive Architectures: CHREST model of human attention and learning

  • Machine Learning: applications (data mining) and theory

  • Methodology: the development of scientific software and models

Current projects

  • "Genetically evolving models of science (GEMS)", an ERC-funded project, with Prof. Fernand Gobet (PI). This five-year project began in October 2019, and is based at the London School of Economics.

  • "Rapid bacteria colony counting algorithm", a HKEP-funded project, with Dr. Na Helian (PI), Dr. Yi Sun and Synoptics Ltd. This four-year project sponsors a PhD student to develop new algorithms for counting bacteria colonies.

Previous projects

  • CHREST - Cognitive architecture
    16th December, 2019
    A cognitive architecture implements a theory of how the human mind works as a computer program, and is used to develop models of behaviour in individual tasks.

  • Reproducible science
    9th May, 2018
    A theme I have returned to several times over the years has been how to improve the reproducibility (or replicability) of science; in particular, the software models used to analyse or model data.

  • Generating scientific theories
    28th February, 2015
    This project combines psychology and computational techniques in an attempt to (semi-)automate the construction of computational models from data.

  • Ferret and n-gram analysis
    15th December 2010
    Ferret is a copy-detection program, locating duplicate text or code in multiple text documents or source files: it is designed to detect copying (collusion) within a given set of files.

  • k-SAT landscape analysis
    finished 2010
    Empirical analysis of landscapes created by k-SAT instances to support a mathematical derivation of the number of local maxima and upper bounds.

  • Simple Synchrony Networks
    finished 2000
    During my PhD I worked on a new kind of neural network and empirically tested its ability to learn to parse English sentences drawn from a corpus of naturally occurring text.