Background
- Collaboration with Sesame while in grad school (1990-94)
- Started simulations(1995-1998)
- Joined Middlesex (1998)
- Proof of Concept (1996-2000)
- EPSRC Fast Stream grant (2001-2003) Modelling Cell Assemblies
as Neuro-Psychological Phenomena and for Practical Applications
- Learning (2000-2003). Hebbian learning (with a range of variants)
can enable networks to learn CA-categories, and thus categorise.
- Hierarchical Categorisation (overlapping CAs are the basis of
hierarchical categories).
- Applications (categorisation and Information Retrieval)
- Process and Binding (2004-2006). These nets can
dynamically move from state to state depending on
internal factors (topology), and external factors (input).
This is a lot more than categorisation. Binding is a problem
of state size and a typical problem of connectionist systems.
- Counting, with Roman at Cognitive Modelling Conference
- Largely unpublished Variable Binding Work
- EPSRC grant (2006-2009) Natural Language Parsing with Cell Assemblies:
computational linguistics with Attractor Nets.
- Workshop on how to use the CANT model in September
