AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

Progressive Neural Architecture Search

Chenxi Liu, Barret Zoph, Maxim Neumann, Jonathon Shlens, Wei Hua, Li-Jia Li, Li Fei-Fei, Alan Yuille, Jonathan Huang, Kevin Murphy

ECCV 2018 · Architectures and efficiency › Neural architecture search · approach: Bi-level differentiable NAS · 1,988 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
progressive search with learned predictor instead of NAS RL (ablation)PNAS · CIFAR-10 val (top 1)models evaluated: 5808 -> 1160Improved
progressive search instead of NAS, total example costPNAS · CIFAR-10compute: about 8 times faster; 21 times less computeImproved
progressive search instead of random search (ablation)PNAS · CIFAR-10 valtop-M accuracyImproved
use MLP-ensemble instead of RNN-ensemble predictor (ablation)PNAS · CIFAR-10 valsearch efficiency: speedup 1.9 -> 5.0Improved
add more blocks to cells (b=1 to b=5) (ablation)PNASNet · CIFAR-10 testtop-1 error: 4.01 -> 3.41Improved
search beyond 5 blocks up to B=10 (ablation)PNAS · CIFAR-10 valtop-M accuracyImproved
compare PNASNet-5 with NASNet-APNASNet-5 · ImageNet (Mobile)top-1 accuracy: 74.0 -> 74.2Improved
compare PNASNet-5 with NASNet-APNASNet-5 · ImageNet (Large)top-1 accuracy: 82.7 -> 82.9Improved
use PNASNet-5 instead of AmoebaNet-CPNASNet-5 · ImageNet (Mobile)top-1 accuracy: 75.7 -> 74.2Hurt
PNAS cell vs NASNet-A on CIFAR-10PNASNet-5 · CIFAR-10 testtop-1 error: 3.41 vs 3.41No clear effect
CIFAR-10 accuracy as proxy for ImageNet accuracyArchitecture search · CIFAR-10 to ImageNetrank correlation: rho 0.727Improved
ensemble predictors for extrapolation to larger cellsMLP-ensemble · Predictor extrapolationSpearman rank correlation: 0.164 vs 0.113 at b=1; 0.468 vs 0.444 at b=4Mixed

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