Explaining How a Neural Network Play the Go Game and Let People Learn
This addresses the challenge in explainable AI of providing verifiable knowledge for human learning in Go, though it appears incremental as it builds on existing AI models.
The paper tackles the problem of explaining knowledge encoded by a Go-playing AI model to teach human players, by extracting interaction primitives from the value network, and experiments demonstrate its effectiveness.
The AI model has surpassed human players in the game of Go, and it is widely believed that the AI model has encoded new knowledge about the Go game beyond human players. In this way, explaining the knowledge encoded by the AI model and using it to teach human players represent a promising-yet-challenging issue in explainable AI. To this end, mathematical supports are required to ensure that human players can learn accurate and verifiable knowledge, rather than specious intuitive analysis. Thus, in this paper, we extract interaction primitives between stones encoded by the value network for the Go game, so as to enable people to learn from the value network. Experiments show the effectiveness of our method.