evosax: JAX-Based Evolution Strategies
Tired of having to handle asynchronous processes for neuroevolution? Do you want to leverage massive vectorization and high-throughput accelerators for evolution strategies (ES)? evosax allows you to leverage JAX, XLA compilation and auto-vectorization/parallelization to scale ES to your favorite accelerators. In this talk we will get to know the core API and how to solve distributed black-box optimization problems with evolution strategies.
Affiliation: Technical University Berlin
I am a 3rd year PhD student working on Evolutionary Meta-Learning at the Technical University Berlin. My work is funded by the Science of Intelligence Excellence Cluster and supervised by Henning Sprekeler. Previously, I completed a MSc in Computing at Imperial College London, a Data Science MSc at Universitat Pompeu Fabra and an Economics undergraduate at University of Cologne. I also interned at DeepMind (Discovery team) & Accenture and maintain a set of open source tools.