Parsing with CAs Home Page

Natural Language Parsing with Cell Assemblies: computational linguistics with attractor nets

The EPSRC has funded a project to develop an agent that communicates via natural language, senses the environment, and assists the user. This agent is based on fatiguing Leaky Integrate and Fire neurons extending our work on Cell Assemblies (CAs).
The project has a significant engineering component. It is based around three main milestones, each an evolution on the prior:
  1. Build a Natural Language (NL) Parser based on CAs.
  2. Build a vision system based on CAs, and integrate this system with the Language system.
  3. Build an agent for a 3D game that assists a user, receives NL input, senses the environment, and has its own goaling mechanism.
The project members include
The project proposal as submitted to EPSRC is here in pdf, here in ps, and here in tex. The current project plan is here in MS Project.
Various issues have pages associated with them.
Chris Huyck is the principal investigator on this project. The project is based on Chris' earlier work on fatiguing Leaky integrate and Fire neurons and Cell Assemblies. Here are some links to an introduction to Cell Assemblies, and some papers, talks, and other related things .