Learning to cooperate, compete, and communicate
OpenAI Blog
Researchers are studying multiagent competitive environments as a training approach for developing more capable AI systems. These environments provide automatic curriculum learning, where difficulty scales with competitor skill, and create continuous pressure for improvement since no stable equilibrium exists. The approach differs fundamentally from single-agent training and requires substantially more research to master effectively.
Why it matters
Multiagent environments where agents compete for resources are stepping stones on the path to AGI. Multiagent environments have two useful properties: first, there is a natural curriculum—the difficulty of the environment is determined by the skill of your competitors (and if you’re competing against clones of yourself, the environment exactly matches your skill level). Second, a multiagent environment has no stable equilibrium: no matter how smart an agent is, there’s always pressure to get smarter. These environments have a very different feel from traditional environments, and it’ll take a lot more research before we become good at them.
