# Network Sketching: Exploiting Binary Structure in Deep CNNs

arXiv: https://arxiv.org/abs/1706.02021 · HTML: https://mlsanity.com/paper/1706.02021/

## Key facts

- Cited by 22 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 7 Jun 2017.

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

- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [A Comprehensive Overview of Large Language Models](https://mlsanity.com/paper/2307.06435/) (2023)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [LUT-GEMM: Quantized Matrix Multiplication based on LUTs for Efficient Inference in Large-Scale Generative Language Models](https://mlsanity.com/paper/2206.09557/) (2022)
- [ARB-LLM: Alternating Refined Binarizations for Large Language Models](https://mlsanity.com/paper/2410.03129/) (2024)
- [ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization](https://mlsanity.com/paper/2406.05981/) (2024)
- [AlphaTuning: Quantization-Aware Parameter-Efficient Adaptation of Large-Scale Pre-Trained Language Models](https://mlsanity.com/paper/2210.03858/) (2022)
- [SlimmeRF: Slimmable Radiance Fields](https://mlsanity.com/paper/2312.10034/) (2023)
- [HGQ: High Granularity Quantization for Real-time Neural Networks on FPGAs](https://mlsanity.com/paper/2405.00645/) (2024)
- [Model Pruning Based on Quantified Similarity of Feature Maps](https://mlsanity.com/paper/2105.06052/) (2021)
- [Improving Quantization-aware Training of Low-Precision Network via Block Replacement on Full-Precision Counterpart](https://mlsanity.com/paper/2412.15846/) (2024)
- [Neural Network Quantization for Microcontrollers: A Comprehensive Survey of Methods, Platforms, and Applications](https://mlsanity.com/paper/2508.15008/) (2025)
- [CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs](https://mlsanity.com/paper/2606.26650/) (2026)
- [Arrhythmia Classifier Using Convolutional Neural Network with Adaptive Loss-aware Multi-bit Networks Quantization](https://mlsanity.com/paper/2202.12943/) (2022)
- [GradFreeBits: Gradient Free Bit Allocation for Dynamic Low Precision Neural Networks](https://mlsanity.com/paper/2102.09298/) (2021)
- [Efficient Visual Recognition with Deep Neural Networks: A Survey on Recent Advances and New Directions](https://mlsanity.com/paper/2108.13055/) (2021)
- [Q-Rater: Non-Convex Optimization for Post-Training Uniform Quantization](https://mlsanity.com/paper/2105.01868/) (2021)
- [Modulating Regularization Frequency for Efficient Compression-Aware Model Training](https://mlsanity.com/paper/2105.01875/) (2021)
- [Quantized Neural Networks via {-1, +1} Encoding Decomposition and Acceleration](https://mlsanity.com/paper/2106.09886/) (2021)
- [AccEPT: An Acceleration Scheme for Speeding Up Edge Pipeline-parallel Training](https://mlsanity.com/paper/2311.05827/) (2023)
- [Binary Quadratic Quantization: Beyond First-Order Quantization for Real-Valued Matrix Compression](https://mlsanity.com/paper/2510.18650/) (2025)
- [Rethinking Output Alignment For 1-bit Post-Training Quantization of Large Language Models](https://mlsanity.com/paper/2512.21651/) (2025)

## Abstract (arXiv)

Convolutional neural networks (CNNs) with deep architectures have substantially advanced the state-of-the-art in computer vision tasks. However, deep networks are typically resource-intensive and thus difficult to be deployed on mobile devices. Recently, CNNs with binary weights have shown compelling efficiency to the community, whereas the accuracy of such models is usually unsatisfactory in practice. In this paper, we introduce network sketching as a novel technique of pursuing binary-weight CNNs, targeting at more faithful inference and better trade-off for practical applications. Our basic idea is to exploit binary structure directly in pre-trained filter banks and produce binary-weight models via tensor expansion. The whole process can be treated as a coarse-to-fine model approximation, akin to the pencil drawing steps of outlining and shading. To further speedup the generated models, namely the sketches, we also propose an associative implementation of binary tensor convolutions. Experimental results demonstrate that a proper sketch of AlexNet (or ResNet) outperforms the existing binary-weight models by large margins on the ImageNet large scale classification task, while the committed memory for network parameters only exceeds a little.