# AutoSlim: Towards One-Shot Architecture Search for Channel Numbers

arXiv: https://arxiv.org/abs/1903.11728 · HTML: https://mlsanity.com/paper/1903.11728/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Mar 2019.

## Built on by (most cited first, 10 of 10)

- [A Comprehensive Survey on Hardware-Aware Neural Architecture Search](https://mlsanity.com/paper/2101.09336/) (2021)
- [Soft Masking for Cost-Constrained Channel Pruning](https://mlsanity.com/paper/2211.02206/) (2022)
- [Maxwell's Demon at Work: Efficient Pruning by Leveraging Saturation of Neurons](https://mlsanity.com/paper/2403.07688/) (2024)
- [Improved Techniques for Quantizing Deep Networks with Adaptive Bit-Widths](https://mlsanity.com/paper/2103.01435/) (2021)
- [CE-NAS: An End-to-End Carbon-Efficient Neural Architecture Search Framework](https://mlsanity.com/paper/2406.01414/) (2024)
- [An Analysis of Super-Net Heuristics in Weight-Sharing NAS](https://mlsanity.com/paper/2110.01154/) (2021)
- [Searching for TrioNet: Combining Convolution with Local and Global Self-Attention](https://mlsanity.com/paper/2111.07547/) (2021)
- [Structural Dropout for Model Width Compression](https://mlsanity.com/paper/2205.06906/) (2022)
- [Multi-Objective Neural Architecture Search by Learning Search Space Partitions](https://mlsanity.com/paper/2406.00291/) (2024)
- [Architecture Aware Latency Constrained Sparse Neural Networks](https://mlsanity.com/paper/2109.00170/) (2021)

## Abstract (arXiv)

We study how to set channel numbers in a neural network to achieve better accuracy under constrained resources (e.g., FLOPs, latency, memory footprint or model size). A simple and one-shot solution, named AutoSlim, is presented. Instead of training many network samples and searching with reinforcement learning, we train a single slimmable network to approximate the network accuracy of different channel configurations. We then iteratively evaluate the trained slimmable model and greedily slim the layer with minimal accuracy drop. By this single pass, we can obtain the optimized channel configurations under different resource constraints. We present experiments with MobileNet v1, MobileNet v2, ResNet-50 and RL-searched MNasNet on ImageNet classification. We show significant improvements over their default channel configurations. We also achieve better accuracy than recent channel pruning methods and neural architecture search methods. Notably, by setting optimized channel numbers, our AutoSlim-MobileNet-v2 at 305M FLOPs achieves 74.2% top-1 accuracy, 2.4% better than default MobileNet-v2 (301M FLOPs), and even 0.2% better than RL-searched MNasNet (317M FLOPs). Our AutoSlim-ResNet-50 at 570M FLOPs, without depthwise convolutions, achieves 1.3% better accuracy than MobileNet-v1 (569M FLOPs). Code and models will be available at: https://github.com/JiahuiYu/slimmable_networks