Recent advances in evolutionary algorithms with examples from neuro-evolution.

J.-B. Mouret

Université Pierre & Marie Curie / CNRS, France

Evolutionary algorithms have come a long way since their birth in the sixties. In this talk, we will highlight a few recent techniques that can be useful to design, optimize and improve models in computational neuroscience. We will give an overview of Pareto-based multi-objective optimization (finding Pareto-optimal trade-offs), neuro-evolution (automatically designing the structure of neural networks), novelty search (avoiding objectives) and multi-objectivization (adding objectives to improve the search).

EvoNeuro: AbstractMouret (last edited 2012-07-06 08:57:32 by BenoitGirard)