SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver · HackerPapers discussions

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver — mentioned by bmc7505

Fri, 23 Dec 2022 09:35:26 GMT

There are architectures with layers which can approximate MAXSAT [1] (or rather, an SDP relaxation thereof), but I doubt this is what OP was referring to. [1]: https://arxiv.org/abs/1905.12149

Paper and discussion history. A mention is not an endorsement.

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver — mentioned by AlexCoventry

Sun, 28 Jun 2020 04:39:18 GMT

SATnet is an interesting paper, in this regard. It's built on a theory of backpropping through the semi-definite programming relaxation of a MAXSAT problem, to learn the MAXSAT problem associated with a class of training examples. It gets good results: learns to solve sudoku puzzles with only binary feedback on its performance (i.e. "that's the solution" or "that's not the solution", no partial feedback on partially-correct solutions). They say it's the first time a neural net has been able to do that. On the other hand, if you look at the coefficient matrix they learn for the MAXSAT problem…

Paper and discussion history. A mention is not an endorsement.

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver — mentioned by bglazer

Wed, 27 May 2020 01:15:15 GMT

You might be interested in the line of work around differentiable relaxations of formal solvers that was kicked off by Zico Kolter et al. They add a Max=SAT solver as a differentiable layer. This allows the neat trick of solving handwritten Sudoku puzzles by composing a ConvNet with the differentiable Max-SAT layer. [1] https://arxiv.org/abs/1905.12149

Paper and discussion history. A mention is not an endorsement.

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver — mentioned by skdotdan

Sun, 29 Sep 2019 18:47:53 GMT

https://arxiv.org/abs/1905.12149

Paper and discussion history. A mention is not an endorsement.

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver — HN submission

Mon, 03 Jun 2019 20:55:28 GMT

SATNet: Bridging deep learning and logical reasoning with differentiable SAT

116 points; 17 discussion replies on HN. Paper and source history. A submission is not an endorsement.