# A Quantization-Friendly Separable Convolution for MobileNets

arXiv: https://arxiv.org/abs/1803.08607 · HTML: https://mlsanity.com/paper/1803.08607/

## Key facts

- Cited by 22 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 22 Mar 2018.

## Built on by (most cited first, 22 of 22)

- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [Pruning and Quantization for Deep Neural Network Acceleration: A Survey](https://mlsanity.com/paper/2101.09671/) (2021)
- [A Survey on Green Deep Learning](https://mlsanity.com/paper/2111.05193/) (2021)
- [Extremely Lightweight Quantization Robust Real-Time Single-Image Super Resolution for Mobile Devices](https://mlsanity.com/paper/2105.10288/) (2021)
- [Make RepVGG Greater Again: A Quantization-aware Approach](https://mlsanity.com/paper/2212.01593/) (2022)
- [Efficient Quantization Strategies for Latent Diffusion Models](https://mlsanity.com/paper/2312.05431/) (2023)
- [Dynamic Precision Analog Computing for Neural Networks](https://mlsanity.com/paper/2102.06365/) (2021)
- [Do All MobileNets Quantize Poorly? Gaining Insights into the Effect of Quantization on Depthwise Separable Convolutional Networks Through the Eyes of Multi-scale Distributional Dynamics](https://mlsanity.com/paper/2104.11849/) (2021)
- [Synergistic Self-supervised and Quantization Learning](https://mlsanity.com/paper/2207.05432/) (2022)
- [Adaptive Precision Training (AdaPT): A dynamic fixed point quantized training approach for DNNs](https://mlsanity.com/paper/2107.13490/) (2021)
- [REx: Data-Free Residual Quantization Error Expansion](https://mlsanity.com/paper/2203.14645/) (2022)
- [MASQuant: Modality-Aware Smoothing Quantization for Multimodal Large Language Models](https://mlsanity.com/paper/2603.04800/) (2026)
- [Rethinking Floating Point Overheads for Mixed Precision DNN Accelerators](https://mlsanity.com/paper/2101.11748/) (2021)
- [IMDeception: Grouped Information Distilling Super-Resolution Network](https://mlsanity.com/paper/2204.11463/) (2022)
- [Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained Processors](https://mlsanity.com/paper/2201.11651/) (2022)
- [A High-Performance Adaptive Quantization Approach for Edge CNN Applications](https://mlsanity.com/paper/2107.08382/) (2021)
- [Dynamic Neural Network Architectural and Topological Adaptation and Related Methods -- A Survey](https://mlsanity.com/paper/2108.10066/) (2021)
- [Designing strong baselines for ternary neural network quantization through support and mass equalization](https://mlsanity.com/paper/2306.17442/) (2023)
- [PIPE : Parallelized Inference Through Post-Training Quantization Ensembling of Residual Expansions](https://mlsanity.com/paper/2311.15806/) (2023)
- [Predicting High-precision Depth on Low-Precision Devices Using 2D Hilbert Curves](https://mlsanity.com/paper/2405.14024/) (2024)
- [Nearly Lossless Adaptive Bit Switching](https://mlsanity.com/paper/2502.01199/) (2025)
- [BD-Net: Has Depth-Wise Convolution Ever Been Applied in Binary Neural Networks?](https://mlsanity.com/paper/2511.17633/) (2025)

## Abstract (arXiv)

As deep learning (DL) is being rapidly pushed to edge computing, researchers invented various ways to make inference computation more efficient on mobile/IoT devices, such as network pruning, parameter compression, and etc. Quantization, as one of the key approaches, can effectively offload GPU, and make it possible to deploy DL on fixed-point pipeline. Unfortunately, not all existing networks design are friendly to quantization. For example, the popular lightweight MobileNetV1, while it successfully reduces parameter size and computation latency with separable convolution, our experiment shows its quantized models have large accuracy gap against its float point models. To resolve this, we analyzed the root cause of quantization loss and proposed a quantization-friendly separable convolution architecture. By evaluating the image classification task on ImageNet2012 dataset, our modified MobileNetV1 model can archive 8-bit inference top-1 accuracy in 68.03%, almost closed the gap to the float pipeline.