# Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution

Source: [Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution](<https://www.sota2.com/research/paper/68fa6a581e209131cea27992>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:22:37.170Z

Authors: Zhang Zhaoyang, Shao Wenqi, Gu Jinwei, Wang Xiaogang, Luo Ping

Published: 2021-06-04T07:10:41.000Z

## Abstract

Model quantization is challenging due to many tedious hyper-parameters such as precision (bitwidth), dynamic range (minimum and maximum discrete values) and stepsize (interval between discrete values). Unlike prior arts that carefully tune these values, we present a fully differentiable approach to learn all of them, named Differentiable Dynamic Quantization (DDQ), which has several benefits. (1) DDQ is able to quantize challenging lightweight architectures like MobileNets, where different layers prefer different quantization parameters. (2) DDQ is hardware-friendly and can be easily implemented using low-precision matrix-vector multiplication, making it capable in many hardware such as ARM. (3) Extensive experiments show that DDQ outperforms prior arts on many networks and benchmarks, especially when models are already efficient and compact. e.g., DDQ is the first approach that achieves lossless 4-bit quantization for MobileNetV2 on ImageNet.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2106.02295>)

[PDF](<https://arxiv.org/pdf/2106.02295.pdf>)