PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Slightly beyond Turing’s computability for studying genetic programming
Olivier Teytaud
In: MCU07, 2007, Orléans.

Abstract

Abstract. Inspired by genetic programming (GP), we study iterative algorithms for non-computable tasks and compare them to naive models. This framework justifies many practical standard tricks from GP and also provides complexity lower-bounds which justify the computational cost of GP thanks to the use of Kolmogorov’s complexity in bounded time.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
ID Code:3196
Deposited By:Olivier Teytaud
Deposited On:20 January 2008