AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

AMC: AutoML for Model Compression and Acceleration on Mobile Devices

Yihui He, Ji Lin, Zhijian Liu, Hanrui Wang, Li-Jia Li, Song Han

ECCV 2018 · Architectures and efficiency › Pruning and efficient tuning · approach: Joint pruning and quantization · 1,477 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
learned RL compression policy instead of uniform handcrafted ratioschannel pruning, ResNet-56 · CIFAR-10 (50% FLOPs)accuracy after fine-tuning (%): 89.8 -> 91.9Improved
learned RL policy instead of best handcrafted policychannel pruning, Plain-20 · CIFAR-10 (50% FLOPs)accuracy before fine-tuning (%): 83.9 -> 86.0Improved
learned RL policy instead of CP expert-tuned ratiosVGG-16 channel pruning · ImageNet (20% FLOPs)top-1 accuracy change (%): -1.7 -> -1.4Improved
learned RL policy instead of uniform width multiplierMobileNet · ImageNet (50% FLOPs)top-1 accuracy change (%): -2.5 -> -0.4Improved
learned RL policy instead of uniform width multiplierMobileNet-V2 · ImageNet (70% FLOPs)top-1 accuracy change (%): -2.0 -> -1.0Improved
use latency-constrained search instead of FLOPs-constrainedAMC on MobileNet · Pixel 1 Android latencyspeedup: 1.81x -> 1.95xImproved
use latency-constrained search instead of FLOPs-constrainedAMC on MobileNet · Titan XP GPUspeedup: 1.43x -> 1.53xImproved
AMC pruning instead of handcrafted pruningMobileNet · ImageNet valtop-1 accuracy (%): 67.2 -> 70.5Improved
AMC policy instead of NetAdapt heuristicMobileNet · Pixel 1 latencylatency (ms): 70.0ms -> 63.3msImproved
iterative fine-grained pruning with AMC instead of expert ratiosResNet-50 · ImageNet valcompression ratio: 3.4x -> 5xImproved
AMC compression of VGG-16 backbone instead of handcrafted 4x pruningFaster R-CNN VGG16 · PASCAL VOC 2007mAP [0.5, 0.95] (%): 36.5 -> 37.2Improved
AMC compression instead of no compressionFaster R-CNN VGG16 · PASCAL VOC 2007mAP (%): 68.7 -> 68.8Improved

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