Evolution strategies as a scalable alternative to reinforcement learning
OpenAI Blog
March 24, 2017
We’ve discovered that evolution strategies (ES), an optimization technique that’s been known for decades, rivals the performance of standard reinforcement learning (RL) techniques on modern RL benchmarks (e.g. Atari/MuJoCo), while overcoming many of RL’s inconveniences.
Verticals
airesearch
Originally published on OpenAI Blog on 3/24/2017