AI is a very big topic, so this column will hit only a few high points. I am indebted to Mark Kantrowitz and Barry Margolin for the information below, which was extracted largely from the AI, Lisp, and Scheme FAQs. Please note that most of these collections will be mirrored in Mark Kantrowitz' yeoman effort, the Carnegie Mellon University Artificial Intelligence Repository (CMU AIR).

The AIR is intended to be a definitive and comprehensive snapshot of worldwide AI freeware. At this writing, the language section (lang) is finished and the remainder is still under development. FTP over and take a look...

Artificial Life

The UCLA Artificial Life Depository contains materials (articles, technical reports, program source code, ...) of interest to Artificial Life (aka ALIFE) researchers. Other artificial life information is available

Fuzzy Logic

The Fuzzy Logic Repository contains information concerning fuzzy logic, including bibliographies, product descriptions and demo versions, machine readable published papers, miscellaneous information, documents and reports, and programs, code and compilers. Start with the INDEX file.

Genetic Algorithms

The Genetic Algorithms Repository includes past copies of the genetic algorithms digest (contact to subscribe) and a copy of Nici Schraudolph's survey of free and commercial GA software. It also contains some software, including GAC (a simple GA written in C), GAL (a simple GA written in Common Lisp), GAucsd, GECO (a Common Lisp toolbox for constructing genetic algorithms), GENESIS, GENOCOP, Paragenesis (a parallel version of GENESIS that runs on the CM-200), and SGA-C (a C implementation/extension of David Goldberg's SGA system).

Genetic Programming

The Genetic Programming Repository contains the archives of the genetic programming mailing list (including the GP FAQ posting), papers and source code. The source code includes the GP implementation from "Genetic Programming" (John Koza, MIT Press, 1992, ISBN 0-262-11170-5) and some related systems. Some of the GP packages available include GPQUICK, Gepetto, GPCplus, and SGPC.

Machine Learning

The UC/Irvine (UCI) AI/Machine Learning Repository has a variety of AI-related materials, with a special focus on machine learning. The directory contains over 80 benchmark data sets for classifier systems (30 MB).

Raymond Mooney, of the Machine Learning Group at the University of Texas at Austin, archives a number of machine learning programs and publications. The archive contains algorithms, code, course materials, papers, and so forth.

The Special Interest Group on Machine Learning of the German Society for Computer Science (GI e.V.) maintains a library of PROLOG implementations of Machine Learning algorithms. The library includes PROLOG implementations of Pat Winston's arch, Jeffrey Becker's AQ-PROLOG, D.H. Fisher's COBWEB, Pavel Brazdil's generation of discriminations from derivation trees, J. Ross Quinlan's ID3, inverse resolution, and Tom Mitchell's version spaces algorithm. The archive is located in

Natural Language Processing

The Consortium for Lexical Research maintains an archive of programs and data files related to natural language processing research, with a particular focus on lexical research. The files 00README.* describe the archive and the consortium itself.

Neural networks

The CMU Simulator Collection specializes in neural networks and connectionist simulators. The archive includes Lisp and C implementations of several of Scott Fahlman's algorithms. It also includes Aspirin/Migraines and Tesauro.

Ohio State University maintains an archive of "neuroprose" (technical reports on connectionist and neural network topics). The archive is located in

Scheme, Lisp, etc.

The Scheme Repository contains a Scheme bibliography, copies of the R4RS report, IEEE P1178 specification and other papers, sample Scheme code for a variety of purposes, several utilities, and some implementations. The Scheme code includes code for calendar calculations, Earley parser, FORMAT for Scheme, a scheme version of the Gabriel benchmarks, Marc Feeley's minimal object support for Scheme, a Scheme pretty-printer, a Prolog interpreter written in Scheme, a random number generator in Scheme, an implementation of SCOOPS, code from "Structure and Interpretation of Computer Programs" (Harold Abelson and Gerald Jay Sussman with Julie Sussman, MIT Press, 1985, ISBN 0-262-01077-1), Aubrey Jaffer's IEEE/R4RS compliance test, an implementation of matrices, a Scheme implementation of the Common Lisp FORMAT function, a Scheme front end to Adobe Illustrator PostScript, and a LALR(1) parser (ZEBU). The repository is located in

The Lisp Repository is part of the CMU AIR (see above), but it started before the AIR, and there's no similar resource. It is located in