AI Paper Atlas

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

Branch & Learn for Recursively and Iteratively Solvable Problems in Predict+Optimize

Xinyi HU, Jasper C.H. Lee, Jimmy H.M. Lee, Allen Z. Zhong

NeurIPS 2022 · Learning theory and algorithms › Combinatorial and search optimization · approach: Unsupervised neural combinatorial optimization · 0 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
B&L vs linear regressionMCFP · USANet real-life, n=100mean regret: 177.55±154.53 -> 141.37±128.52Improved
B&L vs linear regressionMCFP · GEANT artificial, n=100mean regret: 765.13±345.07 -> 733.78±294.23Improved
B&L vs SPO ForestMCFP · GEANT real-life, n=300mean regret: 132.07±77.23 -> 68.44±50.76Improved
B&L vs SPOTMCFP · GEANT real-life, n=100mean regret: 143.57±107.88 -> 72.43±63.58Improved
B&L vs linear regressionMCVC · POLSKA real-life, n=100mean regret: 3.56±2.07 -> 2.22±1.36Improved
B&L vs linear regressionMCVC · PDH artificial, n=300mean regret: 56.59±6.70 -> 53.71±5.62Improved
B&L vs SPOMCVC · PDH artificial, n=300mean regret: 69.46±6.63 -> 53.71±5.62Improved
Increase training set size from 100 to 300B&L · MCFP USANet real-lifemean regret: 141.37±128.52 -> 122.66±85.52Improved
Increase training set size from 100 to 300B&L · MCFP USANet artificialmean regret: 1687.10±729.36 -> 1699.24±688.96No clear effect
Artificial (nonlinear) vs real-life dataB&L · MCFP relative errorrelative error: 7.46%-15.68% -> 1.12%-2.07%Improved
B&L vs SPOT on larger instancesMCFP GEANT · real-life, n=300runtime (s): 439.19 -> 59.17Improved
B&L vs SPO Forest on larger instancesMCVC POLSKA · real-life, n=300runtime (s): 4277.86 -> 1965.00Improved

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