AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Generalized Majorization-Minimization

Sobhan Naderi Parizi, Kun He, Reza Aghajani, Stan Sclaroff, Pedro Felzenszwalb

ICML 2019 · Optimization › Nonconvex and smooth optimization · approach: Coordinate and Adam convergence analysis

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
replace the touching-bound requirement with selection among valid boundsG-MM versus MM on k-means clustering · GMM-200 and Cloudobjective: GMM-200 random partition: 11.20 \pm 0.63 -> 1.85 \pm 0.02; Cloud random partition: 44453 \pm 88341 -> 1470 \pm 8Improved
replace CCP bound selection with random valid-bound selectionLS-SVM on mammals · Mammals, top-left initializationtest error: 42.5 \pm 4.6 -> 31.4 \pm 10.1Improved
select bounds using a multi-fold-MIL-inspired biasG-MM on LS-SVM · Mammals, top-left initializationtest error: 42.5 \pm 4.6 -> 18.9 \pm 5.0Improved
favor coherent part assignments when selecting boundsG-MM on LS-SVM · MIT-Indoor, λ=1.00 initializationaccuracy: 46.62 -> 56.03Improved
use random valid bounds instead of the CCP boundG-MM on LS-SVM · MIT-Indoor, random initializationaccuracy: 41.94 \pm 1.1 -> 47.51 \pm 0.7Improved
increase the progress coefficientG-MM clustering · D31 and Cloudobjective quality across runsHurt
set the progress coefficient to 1, recovering standard k-means/MM bound selection (ablation)G-MM clustering · clustering datasetsobjectiveHurt
use random partition initialization rather than Forgy or k-means++ initialization (ablation)G-MM clustering · D31 and GMM-200objective: GMM-200: 1.85 \pm 0.02; D31: 1.21 \pm 0.05Improved
use coherent rather than less coherent initial part assignments (ablation)LS-SVM on MIT-Indoor · MIT-Indooraccuracy: CCP: 40.88 -> 46.62; G-MM random: 43.38 -> 49.88; G-MM biased: 44.83 -> 56.03Mixed

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