ALL INTELLIGENCE
OpenAI 官方博客T1·Models

OpenAI Releases Evolved Policy Gradients: Auto-Evolving Loss Functions for Faster Training

Original title: Evolved Policy Gradients
48AI Score

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.

Full Report

We’re releasing an experimental metalearning approach called Evolved Policy Gradients, a method that evolves the loss function of learning agents, which can enable fast training on novel tasks. Agents trained with EPG can succeed at basic tasks at test time that were outside their training regime, like learning to navigate to an object on a different side of the room from where it was placed during training.