# ReLeQ: A Reinforcement Learning Approach for Deep Quantization of Neural Networks

arXiv: https://arxiv.org/abs/1811.01704 · HTML: https://mlsanity.com/paper/1811.01704/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Nov 2018.

## Built on by (most cited first, 14 of 14)

- [Mixed-Precision Neural Networks: A Survey](https://mlsanity.com/paper/2208.06064/) (2022)
- [MetaMix: Meta-state Precision Searcher for Mixed-precision Activation Quantization](https://mlsanity.com/paper/2311.06798/) (2023)
- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search](https://mlsanity.com/paper/2308.03290/) (2023)
- [Edge Inference with Fully Differentiable Quantized Mixed Precision Neural Networks](https://mlsanity.com/paper/2206.07741/) (2022)
- [Towards Optimal Compression: Joint Pruning and Quantization](https://mlsanity.com/paper/2302.07612/) (2023)
- [Learnable Mixed-precision and Dimension Reduction Co-design for Low-storage Activation](https://mlsanity.com/paper/2207.07931/) (2022)
- [SimQ-NAS: Simultaneous Quantization Policy and Neural Architecture Search](https://mlsanity.com/paper/2312.13301/) (2023)
- [From Algorithm to Hardware: A Survey on Efficient and Safe Deployment of Deep Neural Networks](https://mlsanity.com/paper/2405.06038/) (2024)
- [ARQ: A Mixed-Precision Quantization Framework for Accurate and Certifiably Robust DNNs](https://mlsanity.com/paper/2410.24214/) (2024)
- [SigmaQuant: Hardware-Aware Heterogeneous Quantization Method for Edge DNN Inference](https://mlsanity.com/paper/2602.22136/) (2026)
- [Modulating Regularization Frequency for Efficient Compression-Aware Model Training](https://mlsanity.com/paper/2105.01875/) (2021)
- [CSMPQ:Class Separability Based Mixed-Precision Quantization](https://mlsanity.com/paper/2212.10220/) (2022)
- [Scale When Needed: Adaptive Neuron-level Mixed Precision Quantization Aware Training](https://mlsanity.com/paper/2605.25054/) (2026)

## Abstract (arXiv)

Deep Neural Networks (DNNs) typically require massive amount of computation resource in inference tasks for computer vision applications. Quantization can significantly reduce DNN computation and storage by decreasing the bitwidth of network encodings. Recent research affirms that carefully selecting the quantization levels for each layer can preserve the accuracy while pushing the bitwidth below eight bits. However, without arduous manual effort, this deep quantization can lead to significant accuracy loss, leaving it in a position of questionable utility. As such, deep quantization opens a large hyper-parameter space (bitwidth of the layers), the exploration of which is a major challenge. We propose a systematic approach to tackle this problem, by automating the process of discovering the quantization levels through an end-to-end deep reinforcement learning framework (ReLeQ). We adapt policy optimization methods to the problem of quantization, and focus on finding the best design decisions in choosing the state and action spaces, network architecture and training framework, as well as the tuning of various hyperparamters. We show how ReLeQ can balance speed and quality, and provide an asymmetric general solution for quantization of a large variety of deep networks (AlexNet, CIFAR-10, LeNet, MobileNet-V1, ResNet-20, SVHN, and VGG-11) that virtually preserves the accuracy (=< 0.3% loss) while minimizing the computation and storage cost. With these DNNs, ReLeQ enables conventional hardware to achieve 2.2x speedup over 8-bit execution. Similarly, a custom DNN accelerator achieves 2.0x speedup and energy reduction compared to 8-bit runs. These encouraging results mark ReLeQ as the initial step towards automating the deep quantization of neural networks.