Accelerating Self-Play Learning in Go
This work addresses the challenge of making learning in large state spaces like Go feasible without extensive computational resources, representing an incremental advancement over existing methods.
The paper tackles the problem of accelerating self-play learning in Go by introducing improvements to the AlphaZero process and architecture, achieving a 50x reduction in computation and surpassing a previous model's performance in 19 days on fewer than 30 GPUs.
By introducing several improvements to the AlphaZero process and architecture, we greatly accelerate self-play learning in Go, achieving a 50x reduction in computation over comparable methods. Like AlphaZero and replications such as ELF OpenGo and Leela Zero, our bot KataGo only learns from neural-net-guided Monte Carlo tree search self-play. But whereas AlphaZero required thousands of TPUs over several days and ELF required thousands of GPUs over two weeks, KataGo surpasses ELF's final model after only 19 days on fewer than 30 GPUs. Much of the speedup involves non-domain-specific improvements that might directly transfer to other problems. Further gains from domain-specific techniques reveal the remaining efficiency gap between the best methods and purely general methods such as AlphaZero. Our work is a step towards making learning in state spaces as large as Go possible without large-scale computational resources.