AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Deep Statistical Solvers

Balthazar Donon, Zhengying Liu, Wenzhuo LIU, Isabelle Guyon, Antoine Marot, Marc Schoenauer

NeurIPS 2020 · Learning theory and algorithms › Combinatorial and search optimization · approach: Constrained neural network optimization · 7 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
use DSS instead of iterative BGSDSS · Poisson linear systemstime per instance (ms): 2.3 -> 1.8Improved
use DSS instead of iterative BGSDSS · Poisson linear systemsloss 50th percentile: 1.7e-2 -> 1.2e-3Improved
train with the problem loss (DSS) instead of distance to ground truth (proxy)DSS · Poisson linear systemsloss 50th percentile: 1.6e-2 -> 1.2e-3Improved
train with the problem loss (DSS) instead of distance to ground truth (proxy)DSS · Poisson linear systemsNRMSE w/ LU: 1.1e-3 -> 1.6e-3Hurt
use DSS instead of Newton-RaphsonDSS · AC power flow IEEE 118time per instance (ms): 2e1 -> 2e-1Improved
use DSS instead of Newton-RaphsonDSS · AC power flow IEEE 14time per instance (ms): 2e1 -> 1e-2Improved
use DSS instead of Newton-RaphsonDSS · AC power flow IEEE 118loss 50th percentile: 4.2e-14 -> 1.7e-6Hurt
train with the problem loss (DSS) instead of proxyDSS · AC power flow IEEE 118correlation of theta_i with NR: 99.99% vs 81.31%Hurt
train with the problem loss (DSS) instead of proxyDSS · AC power flow IEEE 14NRMSE of theta_i: 1.7e-3 -> 7.1e-3Hurt
train with the problem loss (DSS) instead of proxyDSS · AC power flowcorrelation of Pij and Qij with NR: 99.99%No clear effect
train with the problem loss (DSS) instead of proxyDSS · AC power flow IEEE 118loss 50th percentile: 8.3e-6 -> 1.7e-6Improved
test on meshes of different sizes than training (100 to 1000 nodes vs ~500)DSS · Poisson super-generalizationcorrelation with LU: 99.99%No clear effect

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