AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

BlockQNN: Efficient Block-Wise Neural Network Architecture Generation

Zhao Zhong, Zichen Yang, Boyang Deng, Junjie Yan, Wei Wu, Jing Shao, Cheng‐Lin Liu

TPAMI 2020 · Architectures and efficiency › Neural architecture search · approach: Latency-aware architecture search · 152 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
add Q-learning guided search instead of random search (ablation)BlockQNN · CIFAR-100 early-stop accuracyaccuracy: by over 1%Improved
add Q-learning guided search instead of random search with limited parameters (ablation)BlockQNN-L · CIFAR-100 early-stop accuracyaccuracy: by over 1%Improved
add epsilon-greedy exploitation phase after random exploration (ablation)BlockQNN · CIFAR-100 early-stop accuracymean accuracy: 56% -> 65%Improved
search the combined PCC cell instead of separate ReLU, Conv and BN (ablation)BlockQNN search · CIFAR-100 Q-learningaccuracyImproved
add optimal block connection style instead of sequential stacking (ablation)Block-QNN-S · CIFAR-10error rate: 3.30 -> 3.01Improved
add more filters to Block-QNN-Connection (ablation)Block-QNN-Connection · CIFAR-10error rate: 3.01 -> 2.35Improved
add more filters per conv layer (ablation)Block-QNN-S · CIFAR-10error rate: 3.30 -> 2.80Improved
add more filters per conv layer (ablation)Block-QNN-Faster · CIFAR-10error rate: 3.57 -> 3.15Improved
replace conv with depthwise-separable cellBlockQNN · CIFAR-10error rate: 3.30 -> 2.58Improved
use performance predictor instead of training sampled networksFaster BlockQNN · CIFAR-10error rate and cost: 2.80 -> 3.15; 3 days on 32 GPUs -> 0.8 day on 1 GPUMixed
train on 224 and test on 320 instead of 224Block-QNN-Depthwise · ImageNettop-1 error: 19.00 -> 18.00Improved
use Block-QNN-Depthwise instead of NASNet-A (6 @ 4032) at same 224 trainingBlock-QNN-Depthwise · ImageNettop-1 error: 19.90 -> 19.00Improved

Later papers that compared with this method

1 later accepted paper reports BlockQNN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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