[Uber] EvoGrad: Lightweight library for gradient-based evolution
Sep 4, 2019
Introducing EvoGrad: A Lightweight Library for Gradient-Based Evolution
Alex Gajewski, Jeff Clune, Kenneth O. Stanley, and Joel Lehman
quote
To more easily prototype NES-like algorithms, Uber AI researchers built EvoGrad, a Python library that gives researchers the ability to differentiate through expectations (and nested expectations) of random variables, which is key for estimating NES gradients. The idea is to enable more rapid exploration of variants of NES, similar to how TensorFlow enables deep learning research. We believe there are many interesting algorithms yet to be discovered in this vein, and we hope this library will help to catalyze progress in the machine learning community. Beyond EvoGrad, this blog post also announces Evolvability ES, a new NES-based meta-learning algorithm developed by Uber AI researchers that precipitated the creation of EvoGrad.
← Back to all articles Quick Navigation: Next:[ j ] – Prev:[ k ] – List:[ l ]