AI Paper Atlas

Research map / Optimization / Optimizers

Injecting Logical Constraints into Neural Networks via Straight-Through Estimators

Zhun Yang, Joohyung Lee, Chiyoun Park

ICML 2022 · Optimization › Optimizers · approach: Constrained and non-backprop training · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Train with CL-STE constraints rather than NeurASPCNN · Sudoku (Park 70k, inference trick)whole-board accuracy: 66.5% -> 93.7%Improved
Replace NeurASP with CL-STE constraint lossCNN · Sudoku (Park 70k)training time: 13230 -> 813 minutesImproved
Train with CL-STE instead of SATNetSudoku model · Sudoku (Park 70k)test accuracy: 67.78% -> 93.70%Improved
Train with CL-STE instead of SATNetSudoku model · Sudoku (Palm)test accuracy: 6.76% -> 27.37%Improved
Use CL-STE rather than DeepProbLogMNIST addition classifier · mnistAddtraining time: 2565s -> 22sImproved
Use CL-STE rather than NeurASPMNIST addition classifier · mnistAdd2training time: 1682s -> 92sImproved
Use CL-STE rather than symbolic baselinesMNIST addition classifier · mnistAdd3training completion: timeout -> 402sImproved
Use probability binarization b_p(x) instead of b(x) (ablation)CL-STE with iSTE · member(5)accuracy (%): 51.1$\pm$24.9 -> 95.0$\pm$0.5Improved
Use probability binarization b_p(x) instead of b(x) (ablation)CL-STE with iSTE · member(3)accuracy (%): 73.2$\pm$9.1 -> 95.5$\pm$0.7Improved
Use saturated STE instead of iSTE with b(x) (ablation)CL-STE · member(5)accuracy (%): 51.1$\pm$24.9 -> 88.0$\pm$7.1Improved
Use saturated STE instead of iSTE with b(x) (ablation)CL-STE · member(3)accuracy (%): 73.2$\pm$9.1 -> 83.2$\pm$8.4Improved
Add constraint losses to RRN training while keeping labeled data fixedRRN · Sudoku (Palm)test accuracyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy