Meta-trained agents implement Bayes-optimal agents
Mikulik, Vladimir, Delétang, Grégoire, McGrath, Tom, Genewein, Tim, Martic, Miljan, Legg, Shane, Ortega, Pedro A.
–arXiv.org Artificial Intelligence
Memory-based meta-learning is a powerful technique to build agents that adapt fast to any task within a target distribution. A previous theoretical study has argued that this remarkable performance is because the meta-training protocol incentivises agents to behave Bayes-optimally. We empirically investigate this claim on a number of prediction and bandit tasks. Inspired by ideas from theoretical computer science, we show that meta-learned and Bayes-optimal agents not only behave alike, but they even share a similar computational structure, in the sense that one agent system can approximately simulate the other. Furthermore, we show that Bayes-optimal agents are fixed points of the meta-learning dynamics. Our results suggest that memory-based meta-learning might serve as a general technique for numerically approximating Bayes-optimal agents - that is, even for task distributions for which we currently don't possess tractable models.
arXiv.org Artificial Intelligence
Oct-21-2020
- Country:
- North America > United States (0.28)
- Genre:
- Research Report > New Finding (1.00)
- Industry:
- Health & Medicine (0.46)
- Technology: