AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Detecting Interactions from Neural Networks via Topological Analysis

Zirui Liu, Qingquan Song, Kaixiong Zhou, Ting-Hsiang Wang, Ying Shan, Xia Hu

NeurIPS 2020 · Detection and segmentation › Open-set and few-shot detection · approach: Visual relationship and scene graph detection · 4 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 PID instead of NID for pairwise interaction detectionMLP interaction detector · Synthetic F1-F10 averageAUC: 0.951±7.0e−2 -> 0.957±6.2e−2Improved
Use PID instead of NID on F2MLP interaction detector · Synthetic F2AUC: 0.776±4.3e−2 -> 0.804±5.7e-2Improved
Use PID instead of NID on F4MLP interaction detector · Synthetic F4AUC: 0.916±6.3e−2 -> 0.935±3.9e−2Improved
Use PID instead of NID on F1MLP interaction detector · Synthetic F1AUC: 0.985±6.3e−3 -> 0.986±4.1e−3No clear effect
Use PID instead of NID on F10MLP interaction detector · Synthetic F10AUC: 0.989±3.0e−2 -> 0.987±3.5e−2No clear effect
Use PID instead of tree-based Additive Groves on F5Interaction detector · Synthetic F5AUC: 0.67±5.7e-2 -> 1.0±0.0Improved
Use PID instead of tree-based Additive Groves on F6Interaction detector · Synthetic F6AUC: 0.64±1.4e-2 -> 1.0±0.0Improved
Use PID instead of Additive Groves on F10Interaction detector · Synthetic F10AUC: 1.0±0.0 -> 0.987±3.5e−2No clear effect
Add PID-generated synthetic features to random forestRandom forest · Amazon EmployeeAUC: 0.8378±0.0046 -> 0.8460±0.0079Improved
Add NID-generated synthetic features to random forestRandom forest · Amazon EmployeeAUC: 0.8378±0.0046 -> 0.8321±0.0299No clear effect
Add random synthetic features to random forestRandom forest · Amazon EmployeeAUC: 0.8378±0.0046 -> 0.7780±0.0575Hurt
Add PID features to random forestRandom forest · Higgs BosonAUC: 0.7421±0.0019 -> 0.7422±0.0017No clear effect

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