This is an interactive popular-science explainer that visualizes the fundamentals of reinforcement learning, using Braitenberg vehicles to show how neural networks can learn via weight perturbation instead of backpropagation. It notes that OpenAI used a similar evolution-strategies approach in 2017 to train AI to play Atari games. The piece is educational and contains no new research results or model releases.
Evolutionary Strategies
进化策略
A black-box population-based optimization method using fitness selection, applicable to training neural networks.
Curated Intelligence
OpenAI released an experimental metalearning approach called Evolved Policy Gradients (EPG), which uses evolutionary algorithms to optimize the loss function of learning agents. This method enables fast training on novel tasks and allows agents to generalize to scenarios outside their training distribution, such as navigating to objects in different locations.
OpenAI researchers discovered that Evolution Strategies (ES), a decades-old optimization technique, rivals standard Reinforcement Learning (RL) performance on modern benchmarks like Atari and MuJoCo. ES overcomes many of RL's inconveniences while offering superior scalability for agent training.