AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

Performance-Aware Mutual Knowledge Distillation for Improving Neural Architecture Search

Pengtao Xie, Xuefeng Du

CVPR 2022 · Architectures and efficiency › Neural architecture search · approach: Latency-aware architecture search · 16 citations (OpenAlex)

Check 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
Add performance-aware mutual distillationDarts1st · CIFAR-100test error (%): 20.52±0.31 -> 17.61±0.16Improved
Add performance-aware mutual distillationDarts1st · CIFAR-10test error (%): 3.00±0.14 -> 2.38±0.03Improved
Add performance-aware mutual distillationPdarts · CIFAR-100test error (%): 17.43±0.15 -> 15.82±0.11Improved
Add performance-aware mutual distillationPcdarts · CIFAR-10test error (%): 2.57±0.07 -> 2.46±0.05Improved
Add performance-aware mutual distillationPcdarts · CIFAR-100test error (%): 17.01±0.06 -> 15.89±0.07Improved
Add performance-aware mutual distillationPrdarts · CIFAR-100test error (%): 16.48±0.06 -> 16.05±0.04Improved
Add PAMKDPcdarts · ImageNettop-1 test error (%): 24.2 -> 22.8Improved
Add PAMKDPcdarts · ImageNettop-5 test error (%): 7.3 -> 6.4Improved
Add PAMKDDrnas · NAS-Bench-201 CIFAR-10test accuracy: 94.36±0.00 -> 94.83±0.07Improved
Add PAMKDDrnas · NAS-Bench-201 ImageNet-16-120test accuracy: 46.34±0.00 -> 47.59±0.09Improved
Use performance scrutiny instead of unrestricted MKD (ablation)Darts1st mutual distillation · CIFAR-100test error (%): 18.25±0.10 -> 17.61±0.16Improved
Use performance scrutiny instead of unrestricted KDCL (ablation)Darts1st mutual distillation · CIFAR-10test error (%): 2.68±0.06 -> 2.38±0.03Improved

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