# Universally Quantized Neural Compression

arXiv: https://arxiv.org/abs/2006.09952 · HTML: https://mlsanity.com/paper/2006.09952/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Jun 2020.

## Built on by (most cited first, 30 of 64)

- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [Lossy Compression with Gaussian Diffusion](https://mlsanity.com/paper/2206.08889/) (2022)
- [An Introduction to Neural Data Compression](https://mlsanity.com/paper/2202.06533/) (2022)
- [C3: High-performance and low-complexity neural compression from a single image or video](https://mlsanity.com/paper/2312.02753/) (2023)
- [Joint Global and Local Hierarchical Priors for Learned Image Compression](https://mlsanity.com/paper/2112.04487/) (2021)
- [Lossy Compression for Lossless Prediction](https://mlsanity.com/paper/2106.10800/) (2021)
- [Soft then Hard: Rethinking the Quantization in Neural Image Compression](https://mlsanity.com/paper/2104.05168/) (2021)
- [Lossy Image Compression with Quantized Hierarchical VAEs](https://mlsanity.com/paper/2208.13056/) (2022)
- [Algorithms for the Communication of Samples](https://mlsanity.com/paper/2110.12805/) (2021)
- [QARV: Quantization-Aware ResNet VAE for Lossy Image Compression](https://mlsanity.com/paper/2302.08899/) (2023)
- [NVRC: Neural Video Representation Compression](https://mlsanity.com/paper/2409.07414/) (2024)
- [On the advantages of stochastic encoders](https://mlsanity.com/paper/2102.09270/) (2021)
- [Compression with Bayesian Implicit Neural Representations](https://mlsanity.com/paper/2305.19185/) (2023)
- [Spatial Decomposition and Temporal Fusion based Inter Prediction for Learned Video Compression](https://mlsanity.com/paper/2401.15864/) (2024)
- [LVAC: Learned Volumetric Attribute Compression for Point Clouds using Coordinate Based Networks](https://mlsanity.com/paper/2111.08988/) (2021)
- [Towards Empirical Sandwich Bounds on the Rate-Distortion Function](https://mlsanity.com/paper/2111.12166/) (2021)
- [NVTC: Nonlinear Vector Transform Coding](https://mlsanity.com/paper/2305.16025/) (2023)
- [Learning Cross-Scale Weighted Prediction for Efficient Neural Video Compression](https://mlsanity.com/paper/2112.13309/) (2021)
- [Compression with Exact Error Distribution for Federated Learning](https://mlsanity.com/paper/2310.20682/) (2023)
- [Progressive Compression with Universally Quantized Diffusion Models](https://mlsanity.com/paper/2412.10935/) (2024)
- [GIViC: Generative Implicit Video Compression](https://mlsanity.com/paper/2503.19604/) (2025)
- [Universal Exact Compression of Differentially Private Mechanisms](https://mlsanity.com/paper/2405.20782/) (2024)
- [Generalized Gaussian Model for Learned Image Compression](https://mlsanity.com/paper/2411.19320/) (2024)
- [Bridging the Gap between Gaussian Diffusion Models and Universal Quantization for Image Compression](https://mlsanity.com/paper/2504.02579/) (2025)
- [Versatile Learned Video Compression](https://mlsanity.com/paper/2111.03386/) (2021)
- [QGen: On the Ability to Generalize in Quantization Aware Training](https://mlsanity.com/paper/2404.11769/) (2024)
- [Rate Distortion Characteristic Modeling for Neural Image Compression](https://mlsanity.com/paper/2106.12954/) (2021)
- [Trading Information between Latents in Hierarchical Variational Autoencoders](https://mlsanity.com/paper/2302.04855/) (2023)
- [The Rate-Distortion-Perception Trade-off: The Role of Private Randomness](https://mlsanity.com/paper/2404.01111/) (2024)
- [Multi-Sample Training for Neural Image Compression](https://mlsanity.com/paper/2209.13834/) (2022)

## Abstract (arXiv)

A popular approach to learning encoders for lossy compression is to use additive uniform noise during training as a differentiable approximation to test-time quantization. We demonstrate that a uniform noise channel can also be implemented at test time using universal quantization (Ziv, 1985). This allows us to eliminate the mismatch between training and test phases while maintaining a completely differentiable loss function. Implementing the uniform noise channel is a special case of the more general problem of communicating a sample, which we prove is computationally hard if we do not make assumptions about its distribution. However, the uniform special case is efficient as well as easy to implement and thus of great interest from a practical point of view. Finally, we show that quantization can be obtained as a limiting case of a soft quantizer applied to the uniform noise channel, bridging compression with and without quantization.