AI Paper Atlas

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

Discriminative Training of Conditional Random Fields with Probably Submodular Constraints

Maxim Berman, Matthew B. Blaschko

IJCV 2020 · Learning theory and algorithms › Combinatorial and search optimization · approach: Globally optimal geometric 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
relax constraint set from C1 (definitely submodular) to probably submodular C4 (ablation)probably submodular SSVM · TU Darmstadt cowsglobal and average accuracyImproved
relax constraints from definitely submodular C2 to probably submodular C4 (ablation)SSVM · UCSB retinalIoU (%): 85.0 -> 87.2Improved
relax constraints from C2 to probably submodular C4 (ablation)SSVM · UCSB retinal1-Delta_avg (%): 95.7 -> 96.6Improved
use intermediate probably submodular set C3 instead of C2 (ablation)SSVM · UCSB retinalIoU (%): 85.0 -> 85.2No clear effect
add transductive test-set submodularity constraints (C4bar instead of C4) (ablation)probably submodular SSVM · UCSB retinalIoU (%): 87.2 -> 87.2No clear effect
allow pairwise potentials (C1 instead of unary-only C0) (ablation)SSVM · Mitochondria 3D EMtest IoU (%): 43.9 -> 60.4Improved
use C3 instead of larger set C4 (ablation)probably submodular SSVM · Mitochondria 3D EMtest IoU (%): 56.9 -> 64.5Mixed
train with submodular constraints C4 instead of QPBO approximate inferenceSSVM · UCSB retinal1-Delta_avg (%): 92.4 -> 96.6Improved
use constrained C4 with exact graph cuts instead of truncated graph cuts (TGC)SSVM · scene multi-labeltest accuracy (%): 83.3 -> 90.4Improved
use constrained C4 instead of TGCSSVM · yeast multi-labeltest accuracy (%): 71.8 -> 80.0Improved
add 2-stage pretraining and delayed constraint generation (ablation)constraint generation · TU Darmstadt cowsmargins computed: 102.5 -> 6.5 (x10^6)Improved
add delayed constraint generation (1-pass) (ablation)constraint generation · TU Darmstadt cowstotal constraint generation time: 400 s -> 329 sImproved

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