AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Optimal Parameter and Neuron Pruning for Out-of-Distribution Detection

Chao Chen, Zhihang Fu, Kai Liu, Ze Chen, Mingyuan Tao, Jieping Ye

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 2 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
Prune parameters and neurons with unusually high or near-zero energy-score sensitivityResNet50 · ImageNet-1K, averageFPR95: 58.41 -> 25.93Improved
Add neuron pruning to parameter pruning (ablation)ResNet50 · ImageNet-1K, averageFPR95: 34.00 -> 29.12 -> 25.93Improved
Add ReAct feature rectification to OPNP (ablation)ResNet50 · ImageNet-1K, averageFPR95: 25.93 -> 23.12Improved
Use both parameter and neuron pruning rather than parameter pruning alone (ablation)ViT-B/16 · ImageNet-1K, averageFPR95: 24.49 -> 25.19Mixed
Add OPNP to EnergyResNet50 · CIFAR-10, averageFPR95: 9.14 -> 6.61Improved
Add OPNP to ReActResNet50 · CIFAR-100, averageFPR95: 26.05 -> 23.56Mixed
Use sensitivity-guided parameter pruning rather than random or low-magnitude parameter pruning (ablation)ResNet50 · ImageNet-1K / SUNFPR95: 49.42, 43.36 -> 30.40Improved
Use sensitivity-guided neuron pruning rather than random or low-feature-norm neuron pruning (ablation)ResNet50 · ImageNet-1K / SUNFPR95: 52.44, 47.76 -> 26.67Improved
Apply OPNP to the FC layer rather than an intermediate convolutional layer (ablation)ResNet50 · ImageNet-1K / SUNFPR95: 54.98, 54.30, 46.51, 37.06 -> 18.50Improved
Use FC-only pruning rather than global threshold or layer-wise pruning (ablation)ResNet50 · ImageNet-1K / SUNFPR95: 39.74, 40.48 -> 18.50Improved
Estimate sensitivity from 1% rather than 100% of training samples (ablation)OPP · ImageNet-1K / SUNFPR95: 30.40 -> 32.57Hurt
Use variance rather than mean to measure neuron sensitivity (ablation)OPNP · ImageNet-1K / SUNFPR95: 26.7 -> 46.1Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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