AIMay 27, 2020
Neural heuristics for SAT solving
arXiv:2005.13406v113 citations
Originality Incremental advance
AI Analysis
This work addresses the challenge of optimizing SAT-solving performance for computational logic and AI applications, representing an incremental advancement in heuristic design.
The authors tackled the problem of improving SAT-solving algorithms by using neural graph networks with message-passing and attention mechanisms to enhance branching heuristics, resulting in reported improvements over standard human-designed heuristics.
We use neural graph networks with a message-passing architecture and an attention mechanism to enhance the branching heuristic in two SAT-solving algorithms. We report improvements of learned neural heuristics compared with two standard human-designed heuristics.