Research map / Architectures and efficiency / Neural architecture search
SpArSe: Sparse Architecture Search for CNNs on Resource-Constrained Microcontrollers
Igor Fedorov, Ryan P. Adams, Matthew Mattina, Paul Whatmough
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Optimize CNN architecture jointly with structured pruning for MCU memory constraints | SpArSe · CIFAR10-binary | accuracy: 73.02 -> 73.84 | Improved |
| Optimize CNN architecture jointly with structured pruning for MCU memory constraints | SpArSe · Chars4K-binary | accuracy: 74.28 -> 77.78 | Improved |
| Optimize CNN architecture jointly with structured pruning for MCU memory constraints | SpArSe · MNIST | accuracy: 94.38 -> 98.64 | Improved |
| Optimize CNN architecture jointly with structured pruning for MCU memory constraints | SpArSe · CIFAR10-binary | accuracy: 73.02 -> 70.41 | Hurt |
| Add pruning as part of architecture search (ablation) | SpArSe · MNIST | model parameters: ~80x reduction | Improved |
| Optimize working memory rather than model size alone (ablation) | SpArSe on CIFAR10-binary · CIFAR10-binary | working memory: 14.3 KB -> 1.28 KB | Improved |
| Optimize model size rather than both model size and working memory (ablation) | SpArSe on CIFAR10-binary · CIFAR10-binary | working memory: 1.28 KB -> 14.3 KB | Hurt |
Later papers that compared with this method
1 later accepted paper reports Sparse as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference (NeurIPS 2026 · better on 12 of 13 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
- Constrained deep neural network architecture search for IoT devices accounting for hardware calibration (NeurIPS 2019 · similarity 0.91)
- MCUNet: Tiny Deep Learning on IoT Devices (NeurIPS 2020 · similarity 0.90)
- TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge (ICML 2024 · similarity 0.87)
- AtomNet: Designing Tiny Models from Operators Under Extreme MCU Constraints (AAAI 2025 · similarity 0.86)
- msf-CNN: Patch-based Multi-Stage Fusion with Convolutional Neural Networks for TinyML (NeurIPS 2025 · similarity 0.85)
- TinyTTA: Efficient Test-time Adaptation via Early-exit Ensembles on Edge Devices (NeurIPS 2024 · similarity 0.85)
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.
