AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

An Evaluation of Zero-Cost Proxies - from Neural Architecture Performance Prediction to Model Robustness

Jovita Lukasik, Michael Moeller, Margret Keuper

IJCV 2024 · Architectures and efficiency › Neural architecture search · approach: One-shot and few-shot supernets · 6 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 all zero-cost proxies instead of only the most important one (ablation)random forest predictor · CIFAR-10 PGD robust accuracytest R2: -0.31 -> 0.65Improved
use all zero-cost proxies instead of only the most important one (ablation)random forest predictor · CIFAR-10 clean accuracytest R2: 0.84 -> 0.97Improved
use all zero-cost proxies instead of only the most important one (ablation)random forest predictor · ImageNet16-120 FGSM accuracytest R2: -0.003 -> 0.65Improved
use all zero-cost proxies instead of only the most important one (ablation)random forest predictor · CIFAR-100 clean accuracytest R2: 0.74 -> 0.96Improved
add the most important proxy to the feature set (ablation)random forest multi-objective · CIFAR-10 Clean-FGSMtest R2: 0.92 -> 0.92No clear effect
add the most important proxy to the feature set (ablation)random forest multi-objective · CIFAR-10 Clean-PGDtest R2: 0.79 -> 0.81Improved
add the most important proxy to the feature set (ablation)random forest multi-objective · ImageNet16-120 Clean-FGSMtest R2: 0.72 -> 0.78Improved
add the most important proxy to the feature set (ablation)random forest multi-objective · CIFAR-100 Clean-PGDtest R2: 0.83 -> 0.83No clear effect
use stronger attacks (PGD/APGD instead of FGSM)Kendall tau of ZCPs with robust accuracy · CIFAR-10rank correlationHurt

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