AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

ChamNet: Towards Efficient Network Design Through Platform-Aware Model Adaptation

Xiaoliang Dai, Peizhao Zhang, Bichen Wu, Hongxu Yin, Fei Sun, Yanghan Wang, Marat Dukhan, Yunqing Hu, Yiming Wu, Yangqing Jia, Peter Vajda, Matt Uyttendaele and 1 more

CVPR 2019 · Architectures and efficiency › Neural architecture search · approach: Latency-aware architecture search · 286 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
adapt MobileNetV2 building blocks to fit the target platform and latency budgetChamNet-Mobile · ImageNet, Snapdragon 835 CPU, 4.0 mstop-1 accuracy: 8.5% absolute accuracy gainImproved
adapt MobileNetV2 building blocks to fit the target platform and latency budgetChamNet-Mobile · ImageNet, Snapdragon 835 CPU, 20 mstop-1 accuracy: 73.8%Improved
adapt MobileNetV2 building blocks to fit the target platform and latency budgetChamNet-Mobile · ImageNet, Hexagon v62 DSP, 20 mstop-1 accuracy: 75.3%Improved
adapt MobileNetV2 building blocks for energy-constrained inferenceChamNet-Mobile · ImageNet, Snapdragon 835 CPU, 14mJ per runtop-1 accuracy: 58.8% -> 60.0%Improved
adapt MobileNetV2 building blocks for energy-constrained inferenceChamNet-Mobile · ImageNet, Snapdragon 835 CPUenergy per run: 19mJ -> 14mJImproved
redistribute computation from early stages toward later stagesChamNet-Mobile · ImageNet, Snapdragon 835 CPUtop-1 accuracy: 69.8% -> 71.9%Improved
use direct platform latency measurements rather than FLOPs as the optimization targetChameleon · mobile CPU and DSPlatencyImproved
select accuracy-predictor training samples using Bayesian optimization with exploration and exploitationChameleon accuracy predictor · MobileNetV2 accuracy predictionsample efficiencyImproved
use an operator latency lookup table for candidate latency estimationChameleon latency predictor · target-device network latency estimationestimation time: less than one CPU secondImproved

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