AI Paper Atlas

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

Novel Upper Bounds for the Constrained Most Probable Explanation Task

Tahrima Rahman, Sara Rouhani, Vibhav Giridhar Gogate

NeurIPS 2021 · Learning theory and algorithms › Combinatorial and search optimization · approach: Predict-then-optimize learning · 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
Use MPE-based bound (MB) instead of MCKP-based bound (KP) on tractable cutset networksMB · Cutset networks (AD q20)upper bound: -3307.8 -> -3308.3Improved
Use MB instead of KP on cutset network ReutersMB · Reuters q50upper bound: -1431.6 -> -1443.5Improved
Use KP/JG instead of MB/MM on high-treewidth UAI networks at i-bound 2KP · Segmentation q20upper bound: -381.7 -> -459.9Improved
Use KP/JG instead of MB on Grids at i-bound 2KP · Grids q20upper bound: 1370.4 -> 821.4Improved
Use KP/JG instead of MB/MM on UAI networksKP and JG · UAI networkscomputation time: 1 -> 430Hurt
Increase i-bound from 2 to 10 (ablation)MB · Segmentation q20upper bound: -381.7 -> -452.7Improved
Increase i-bound from 2 to 10 (ablation)MM · Grids q20upper bound: 1040.8 -> 824.1Improved
Increase i-bound from 2 to 10 (ablation)KP · Segmentation q20upper bound: -459.9 -> -460.0No clear effect
Use MM instead of MB at same i-boundMM · Grids q20 iB=2upper bound: 1370.4 -> 1040.8Improved
Increase i-bound from 2 to 4 on MNIST grid model (ablation)MB · MNIST grid modelaverage upper bound: 291.42 -> 290.79Improved
Use JG instead of MB on MNIST grid modelJG · MNIST grid modelaverage upper bound: 291.42 -> 290.11Improved
Use grid distance model instead of univariate distance modelGrid model · MNIST adversarialperturbation smoothnessImproved

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