Recognition: unknown
Learning a SAT Solver from Single-Bit Supervision
read the original abstract
We present NeuroSAT, a message passing neural network that learns to solve SAT problems after only being trained as a classifier to predict satisfiability. Although it is not competitive with state-of-the-art SAT solvers, NeuroSAT can solve problems that are substantially larger and more difficult than it ever saw during training by simply running for more iterations. Moreover, NeuroSAT generalizes to novel distributions; after training only on random SAT problems, at test time it can solve SAT problems encoding graph coloring, clique detection, dominating set, and vertex cover problems, all on a range of distributions over small random graphs.
This paper has not been read by Pith yet.
Forward citations
Cited by 3 Pith papers
-
Q3SAT-GPT: A Generative Model for Discovering Quantum Circuits for the 3-SAT Problem
A generative model learns patterns from adaptive QAOA circuits to generate high-quality shallow quantum circuits for Max-E3-SAT that scale better than variational baselines.
-
One Step Forward and K Steps Back: Better Reasoning with Denoising Recursion Models
Denoising Recursion Models train multi-step noise reversal in looped transformers and outperform the prior Tiny Recursion Model on ARC-AGI.
-
Relational inductive biases, deep learning, and graph networks
Graph networks unify graph-based neural methods into a general framework with strong relational inductive biases to support combinatorial generalization and structured reasoning in AI.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.