# SEAM: Searching Transferable Mixed-Precision Quantization Policy through Large Margin Regularization

arXiv: https://arxiv.org/abs/2302.06845 · HTML: https://mlsanity.com/paper/2302.06845/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 54 references that are in the map.
- First posted on arXiv 14 Feb 2023.

## Builds on (most cited first, 30 of 38)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [Up or Down? Adaptive Rounding for Post-Training Quantization](https://mlsanity.com/paper/2004.10568/) (2020)
- [Rethinking the Value of Network Pruning](https://mlsanity.com/paper/1810.05270/) (2018)
- [DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients](https://mlsanity.com/paper/1606.06160/) (2016)
- [PACT: Parameterized Clipping Activation for Quantized Neural Networks](https://mlsanity.com/paper/1805.06085/) (2018)
- [HAQ: Hardware-Aware Automated Quantization with Mixed Precision](https://mlsanity.com/paper/1811.08886/) (2018)
- [Single Path One-Shot Neural Architecture Search with Uniform Sampling](https://mlsanity.com/paper/1904.00420/) (2019)
- [HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision](https://mlsanity.com/paper/1905.03696/) (2019)
- [Large-Margin Softmax Loss for Convolutional Neural Networks](https://mlsanity.com/paper/1612.02295/) (2016)
- [HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networks](https://mlsanity.com/paper/1911.03852/) (2019)
- [Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks](https://mlsanity.com/paper/1908.05033/) (2019)
- [Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search](https://mlsanity.com/paper/1812.00090/) (2018)
- [Rethinking Differentiable Search for Mixed-Precision Neural Networks](https://mlsanity.com/paper/2004.05795/) (2020)
- [Nonuniform-to-Uniform Quantization: Towards Accurate Quantization via Generalized Straight-Through Estimation](https://mlsanity.com/paper/2111.14826/) (2021)
- [Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimization](https://mlsanity.com/paper/2110.06554/) (2021)
- [FracBits: Mixed Precision Quantization via Fractional Bit-Widths](https://mlsanity.com/paper/2007.02017/) (2020)
- [Mixed-Precision Neural Network Quantization via Learned Layer-wise Importance](https://mlsanity.com/paper/2203.08368/) (2022)
- [PROFIT: A Novel Training Method for sub-4-bit MobileNet Models](https://mlsanity.com/paper/2008.04693/) (2020)
- [Orthogonal Projection Loss](https://mlsanity.com/paper/2103.14021/) (2021)
- [HMQ: Hardware Friendly Mixed Precision Quantization Block for CNNs](https://mlsanity.com/paper/2007.09952/) (2020)
- [Rethinking Feature Distribution for Loss Functions in Image Classification](https://mlsanity.com/paper/1803.02988/) (2018)
- [Search What You Want: Barrier Panelty NAS for Mixed Precision Quantization](https://mlsanity.com/paper/2007.10026/) (2020)

## Built on by (most cited first, 6 of 6)

- [Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision](https://mlsanity.com/paper/2307.00331/) (2023)
- [Retraining-free Model Quantization via One-Shot Weight-Coupling Learning](https://mlsanity.com/paper/2401.01543/) (2024)
- [TMPQ-DM: Joint Timestep Reduction and Quantization Precision Selection for Efficient Diffusion Models](https://mlsanity.com/paper/2404.09532/) (2024)
- [Investigating the Impact of Quantization on Adversarial Robustness](https://mlsanity.com/paper/2404.05639/) (2024)
- [Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligning](https://mlsanity.com/paper/2505.04877/) (2025)
- [Precision Neural Network Quantization via Learnable Adaptive Modules](https://mlsanity.com/paper/2504.17263/) (2025)

## Abstract (arXiv)

Mixed-precision quantization (MPQ) suffers from the time-consuming process of searching the optimal bit-width allocation i.e., the policy) for each layer, especially when using large-scale datasets such as ISLVRC-2012. This limits the practicality of MPQ in real-world deployment scenarios. To address this issue, this paper proposes a novel method for efficiently searching for effective MPQ policies using a small proxy dataset instead of the large-scale dataset used for training the model. Deviating from the established norm of employing a consistent dataset for both model training and MPQ policy search stages, our approach, therefore, yields a substantial enhancement in the efficiency of MPQ exploration. Nonetheless, using discrepant datasets poses challenges in searching for a transferable MPQ policy. Driven by the observation that quantization noise of sub-optimal policy exerts a detrimental influence on the discriminability of feature representations -- manifesting as diminished class margins and ambiguous decision boundaries -- our method aims to identify policies that uphold the discriminative nature of feature representations, i.e., intra-class compactness and inter-class separation. This general and dataset-independent property makes us search for the MPQ policy over a rather small-scale proxy dataset and then the policy can be directly used to quantize the model trained on a large-scale dataset. Our method offers several advantages, including high proxy data utilization, no excessive hyper-parameter tuning, and high searching efficiency. We search high-quality MPQ policies with the proxy dataset that has only 4% of the data scale compared to the large-scale target dataset, achieving the same accuracy as searching directly on the latter, improving MPQ searching efficiency by up to 300 times.