Learning Montezuma's Revenge from a Single Demonstration

Requests for Research 2.0

Evolution Strategies as a Scalable Alternative to Reinforcement Learning

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, while overcoming many of RL's inconveniences.

Generative Models

This post describes four projects that share a common theme of enhancing or using generative models, a branch of unsupervised learning techniques in machine learning.