atlas

Research map / Learning theory and algorithms / Submodular and game algorithms

Decomposable Submodular Function Minimization: Discrete and Continuous

NeurIPS 2017 · Learning theory and algorithms › Submodular and game algorithms · approach: Monotone submodular maximization · 15 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 level-1 IBFS instead of coordinate descentDSFM on image-segmentation instances with small cliques · Small cliquesrunning timeMixed
Use level-1 IBFS instead of gradient methodsDSFM with larger cliques or generic level-0 methods · Large cliquesrunning timeHurt
Use coordinate descent with an approximate Fujishige-Wolfe level-0 solverDSFM with large region potentials · Large cliquesability to obtain an approximate solutionHelped
Warm-start Fujishige-Wolfe from the current solution and run 10 iterations (ablation)Fujishige-Wolfe level-0 solver with coordinate descent · Large cliques with region potentialsrunning timeHelped
Use a generic Fujishige-Wolfe level-0 solver instead of the potential-specific quadratic minimization algorithm (ablation)RCDM and ACDM · Large cliques with region potentialsrunning time: 4.29954 -> 4.4422Hurt

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