AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

MCUNet: Tiny Deep Learning on IoT Devices

Ji Lin, Wei-Ming Chen, Yujun Lin, john cohn, Chuang Gan, Song Han

NeurIPS 2020 · Architectures and efficiency › Neural architecture search · approach: Latency-aware architecture search · 269 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
jointly optimize TinyNAS architecture search and TinyEngine inferenceMCUNet · ImageNet, STM32H743top-1 accuracy: 70.7% vs 69.8%Improved
use TinyEngine rather than CMSIS-NN with scaled MobileNetV2 (ablation)MobileNetV2 · ImageNet, 320kB SRAM and 1MB Flashtop-1 accuracy: 35.2% -> 47.4%Improved
use TinyEngine rather than CMSIS-NN with scaled ProxylessNAS (ablation)ProxylessNAS · ImageNet, 320kB SRAM and 1MB Flashtop-1 accuracy: 49.5% -> 56.4%Improved
add TinyNAS to TinyEngine-based model deployment (ablation)MCUNet · ImageNet, 320kB SRAM and 1MB Flashtop-1 accuracy: 47.4% -> 61.8%; 56.4% -> 61.8%Improved
optimize the search space using satisfying-network FLOPs (ablation)TinyNAS · ImageNet-100top-1 accuracy: 2.3% better accuracyImproved
use a huge variable-resolution and variable-width search space instead of an optimized space (ablation)TinyNAS · ImageNet-100top-1 accuracyHurt
use evolutionary search rather than random search (ablation)TinyNAS with TinyEngine · search accuracybest accuracy: 1%Improved
replace layer-wise scheduling with model-adaptive im2col scheduling (ablation)TinyEngine · inferenceinference efficiency: 13%Improved
specialize computation kernels with tiling, unrolling, and operation fusion (ablation)TinyEngine · inferenceinference efficiency: 22%Improved
use in-place depth-wise convolution (ablation)TinyEngine · inferencememory usage: 1.6×Improved
replace runtime interpretation with code generation (ablation)TinyEngine · inference versus CMSIS-NNmemory usage: 2.1×Improved
replace runtime interpretation with code generation (ablation)TinyEngine · inference versus CMSIS-NNinference efficiency: 22%Improved

Later papers that compared with this method

7 later accepted papers report MCUNetV2 as a baseline in their results tables; 7 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