# Transform Quantization for CNN (Convolutional Neural Network) Compression

arXiv: https://arxiv.org/abs/2009.01174 · HTML: https://mlsanity.com/paper/2009.01174/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Sep 2020.

## Built on by (most cited first, 20 of 20)

- [Towards Efficient Post-training Quantization of Pre-trained Language Models](https://mlsanity.com/paper/2109.15082/) (2021)
- [An Information-Theoretic Justification for Model Pruning](https://mlsanity.com/paper/2102.08329/) (2021)
- [Matrix Compression via Randomized Low Rank and Low Precision Factorization](https://mlsanity.com/paper/2310.11028/) (2023)
- [Knowledge Distillation in Vision Transformers: A Critical Review](https://mlsanity.com/paper/2302.02108/) (2023)
- [KV Cache Transform Coding for Compact Storage in LLM Inference](https://mlsanity.com/paper/2511.01815/) (2025)
- [Design Space Exploration of Low-Bit Quantized Neural Networks for Visual Place Recognition](https://mlsanity.com/paper/2312.09028/) (2023)
- [Neural Network Compression for Noisy Storage Devices](https://mlsanity.com/paper/2102.07725/) (2021)
- [LilNetX: Lightweight Networks with EXtreme Model Compression and Structured Sparsification](https://mlsanity.com/paper/2204.02965/) (2022)
- [Radio: Rate-Distortion Optimization for Large Language Model Compression](https://mlsanity.com/paper/2505.03031/) (2025)
- [Illumination and Temperature-Aware Multispectral Networks for Edge-Computing-Enabled Pedestrian Detection](https://mlsanity.com/paper/2112.05053/) (2021)
- [From Algorithm to Hardware: A Survey on Efficient and Safe Deployment of Deep Neural Networks](https://mlsanity.com/paper/2405.06038/) (2024)
- [Foundations of Large Language Model Compression -- Part 1: Weight Quantization](https://mlsanity.com/paper/2409.02026/) (2024)
- [Minimax Optimal Quantization of Linear Models: Information-Theoretic Limits and Efficient Algorithms](https://mlsanity.com/paper/2202.11277/) (2022)
- [Bandwidth-efficient Inference for Neural Image Compression](https://mlsanity.com/paper/2309.02855/) (2023)
- [A Soft SIMD Based Energy Efficient Computing Microarchitecture](https://mlsanity.com/paper/2212.09358/) (2022)
- [Progressive Ensemble Distillation: Building Ensembles for Efficient Inference](https://mlsanity.com/paper/2302.10093/) (2023)
- [Purification Of Contaminated Convolutional Neural Networks Via Robust Recovery: An Approach with Theoretical Guarantee in One-Hidden-Layer Case](https://mlsanity.com/paper/2407.11031/) (2024)
- [An Efficient 3D Convolutional Neural Network with Channel-wise, Spatial-grouped, and Temporal Convolutions](https://mlsanity.com/paper/2503.00796/) (2025)
- [Delta-SVD: Efficient Compression for Personalized Text-to-Image Models](https://mlsanity.com/paper/2508.16863/) (2025)
- [Securing Radiation Detection Systems with an Efficient TinyML-Based IDS for Edge Devices](https://mlsanity.com/paper/2509.01592/) (2025)

## Abstract (arXiv)

In this paper, we compress convolutional neural network (CNN) weights post-training via transform quantization. Previous CNN quantization techniques tend to ignore the joint statistics of weights and activations, producing sub-optimal CNN performance at a given quantization bit-rate, or consider their joint statistics during training only and do not facilitate efficient compression of already trained CNN models. We optimally transform (decorrelate) and quantize the weights post-training using a rate-distortion framework to improve compression at any given quantization bit-rate. Transform quantization unifies quantization and dimensionality reduction (decorrelation) techniques in a single framework to facilitate low bit-rate compression of CNNs and efficient inference in the transform domain. We first introduce a theory of rate and distortion for CNN quantization, and pose optimum quantization as a rate-distortion optimization problem. We then show that this problem can be solved using optimal bit-depth allocation following decorrelation by the optimal End-to-end Learned Transform (ELT) we derive in this paper. Experiments demonstrate that transform quantization advances the state of the art in CNN compression in both retrained and non-retrained quantization scenarios. In particular, we find that transform quantization with retraining is able to compress CNN models such as AlexNet, ResNet and DenseNet to very low bit-rates (1-2 bits).