Multi-Agent Reinforcement Learning
They’ve taken this approach with Cicero.
Why is MARL hard?
There are many reasons why MARL is hard, including:
- Non-stationarity of the environment, making theoretical analysis super hard
- Joint action space exponentiates search space in size
- “who knows what”?
But MARL is the future. Autocurricula is a big part of it, to see emergent behavior.
From the DeepMind Paper: Current approaches to artificial intelligence (AI) focus too much on individual agents and not enough on the interactions between agents in a multi-agent system. We should focus on “multi-agent intelligence,” studying how agent interactions lead to the emergence of new behaviours and innovations.
A focus on multi-agent intelligence and autocurricula can lead to the development of more robust, adaptable, and innovative AI systems that can better handle complex, real-world problems.