Deep learning for logic optimization

The slowing down of Moore’s law and the emergence of new technologies puts an increasing pressure on the field of EDA in general and the logic synthesis and optimization community in particular. There is a constant need to improve optimization heuristics. However, finding and implementing such heuristics is a difficult task, especially with the novel logic primitives and potentially unconventional requirements of emerging technologies. In this paper, we show how logic optimization may be recast as a game. We then take advantage of recent advances in deep reinforcement learning to build a system that learns how to play this game. Our design has a number of desirable properties. It is autonomous because it learns automatically, and does not require handcrafted heuristics or other human intervention. It generalizes to large multi-output Boolean functions after training on small examples. Additionally, it natively supports both singleand multi-output functions, without the need to handle special cases. Finally, it is generic because the same algorithm can be used to achieve different optimization objectives, eg, size and depth.