Engineering Insights

Competitive self-play

We’ve found that self-play allows simulated AIs to discover physical skills like tackling, ducking, faking, kicking, catching, and diving for the ball, without explicitly designing an environment with these skills in mind. Self-play ensures that the environment is always the right difficulty for an AI to improve. Taken alongside our Dota 2 self-play results, we have increasing confidence that self-play will be a core part of powerful AI systems in the future.

Next.js 4: React 16 and styled-jsx 2

We are happy to introduce , which features support for React 16 and introduces a major upgrade for the default styling engine styled-jsx with support for .Next.js 4dynamic styles Read more

Node.js 6.11.4 (LTS)

Path validation vulnerability, September 2017

Nonlinear computation in deep linear networks

Node.js 8.6.0 (Current)

Learning to model other minds

We’re releasing an algorithm which accounts for the fact that other agents are learning too, and discovers self-interested yet collaborative strategies like tit-for-tat in the iterated prisoner’s dilemma. This algorithm, Learning with Opponent-Learning Awareness (LOLA), is a small step towards agents that model other minds.

Learning with opponent-learning awareness