AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Adaptive Interaction Modeling via Graph Operations Search

Haoxin Li, Wei-Shi Zheng, Yu Tao, Haifeng Hu, Jian-Huang Lai

CVPR 2020 · Graph learning › Knowledge and temporal graphs · approach: Graph reasoning for visual recognition · 5 citations (OpenAlex)

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 searched adaptive structures rather than global poolingI3D-ResNet interaction recognition · Something-Something-V1 Valaccuracy: 48.1 -> 51.4Improved
Use searched adaptive structures rather than global poolingI3D-ResNet interaction recognition · Something-Something-V2 Valaccuracy: 60.3 -> 63.5Improved
Use an adaptive structure per video rather than a single non-adaptive structure at test time (ablation)Graph operations search · Something-Something-V1/V2 Valaccuracy: 50.2 -> 51.4; 62.4 -> 63.5Improved
Add feature aggregation as the graph operation (ablation)I3D-ResNet interaction recognition · Something-Something-V1/V2 Valaccuracy: 48.1 -> 49.9; 60.3 -> 62.0Improved
Add difference propagation as the graph operation (ablation)I3D-ResNet interaction recognition · Something-Something-V1/V2 Valaccuracy: 48.1 -> 49.5; 60.3 -> 61.8Improved
Add temporal convolution as the graph operation (ablation)I3D-ResNet interaction recognition · Something-Something-V1/V2 Valaccuracy: 48.1 -> 48.7; 60.3 -> 61.0Improved
Add background incorporation as the graph operation (ablation)I3D-ResNet interaction recognition · Something-Something-V1/V2 Valaccuracy: 48.1 -> 49.7; 60.3 -> 62.4Improved
Add node attention as the graph operation (ablation)I3D-ResNet interaction recognition · Something-Something-V1/V2 Valaccuracy: 48.1 -> 49.8; 60.3 -> 61.8Improved
Use fixed substructures instead of original graph operations as search primitives (ablation)Graph operations search · Something-Something-V1/V2 Valaccuracy: 51.0 -> 51.4; 63.1 -> 63.5Improved
Add diversity regularization (ablation)Non-adaptive graph operations search · Something-Something-V1 Valaccuracy: 49.9 -> 50.7Improved
Transfer learned adaptive structure weights across datasets (ablation)Adaptive structures search · Something-Something-V1/V2 Valaccuracy: 51.4 -> 51.0; 63.5 -> 63.4No clear effect
Increase the computation cell from 2 to 3 intermediate supernodes (ablation)Adaptive structures search · Something-Something-V1/V2 Valaccuracy: 50.6 -> 51.4; 63.1 -> 63.5Improved

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