FairQuant: Fairness-Aware Mixed-Precision Quantization for Medical Image Classification

Thomas Woergaard; Raghavendra Selvan

cs.CVFeb 26, 2026

FairQuant: Fairness-Aware Mixed-Precision Quantization for Medical Image Classification

Authors: Thomas Woergaard, Raghavendra Selvan

Organizations: Department of Computer Science, University of Copenhagen, Denmark

Abstract

Compressing neural networks by quantizing model parameters offers useful trade-off between performance and efficiency. Methods like quantization-aware training and post-training quantization strive to maintain the downstream performance of compressed models compared to the full precision models. However, these techniques do not explicitly consider the impact on algorithmic fairness. In this work, we study fairness-aware mixed-precision quantization schemes for medical image classification under explicit bit budgets. We introduce FairQuant, a framework that combines group-aware importance analysis, budgeted mixed-precision allocation, and a learnable Bit-Aware Quantization (BAQ) mode that jointly optimizes weights and per-unit bit allocations under bitrate and fairness regularization. We evaluate the method on Fitzpatrick17k and ISIC2019 across ResNet18/50, DeiT-Tiny, and TinyViT. Results show that FairQuant configurations with average precision near 4-6 bits recover much of the Uniform 8-bit accuracy while improving worst-group performance relative to Uniform 4- and 8-bit baselines, with comparable fairness metrics under shared budgets.

Figures & tables

Figure 1: High-level overview of the proposed FairQuant framework (top pathway) illustrating the group importance calculation based on a calibration set, which is then used to allocate appropriate bit widths to the weights. Regular post-training quantization (bottom pathway).

Dataset	Model	Method	AvgBits $\downarrow$	GBOPs $\downarrow$	AvgAcc $\uparrow$	WorstAcc $\uparrow$	Gap $\downarrow$	EOpp0 $\downarrow$	EOdd $\downarrow$
FitzP.	DeiT-Tiny	FP32	32.00	0.07	53.00	47.00	12.10	0.003	0.225
		U8	8.00	0.02	53.00	47.00	12.00	0.003	0.224
		U4	4.00	0.01	45.50	40.20	10.60	0.003	0.222
		Lin et al. [ 16 ]	4.80	0.01	32.60	28.70	7.80	0.004	0.141
		FQ-QAT	5.03	0.01	54.10 ±0.46	47.90 ±0.56	12.30 ±0.72	0.003	0.229
		FQ-BAQ	4.30	0.01	51.83 ±0.64	46.00 ±0.10	11.67 ±1.31	0.003	0.220

Table 1: Main results across two datasets and different models. All QAT experiments use the following bit allocation {20%: 2, 40%: 4, 40%: 8}, and BAQ is reported for range 4-8. Significant results for average bits, average accuracy, and worst group accuracy compared to the baseline methods are shown in bold.

Panel of Figure 2 — Figure 2: Ablations and stability for BAQ. Left: sweep of the bitrate regularizer $\lambda_{\mathrm{baq},b}$ showing test AvgAcc and EOpp on Fitzpatrick17k and ISIC2019. Middle: sweep of the fairness-loss scale $\lambda_{\mathrm{fair}}$ on the ResNet-18 backbone. Right: learning-rate sweep for BAQ bit proxies. All panels use a fixed BAQ bit interval $[b_{\min},b_{\max}]~=[4,16]$ . Shaded regions indicate variability across repeated runs.

Explore similar work

Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization
Oct 7, 2026Samy Houache, Yann Traonmilin, Jean-François AujolPost-Training Quantization Neural Network Quantization
IMB, UB · UB, IMB
Weight Group-wise Post-Training Quantization for Medical Foundation Model
Apr 9, 2026Yineng Chen, Peng Huang, Aozhong Zhang +9Post-Training Quantization Low-Bit Quantization
University at Albany, SUNY · Southwest Jiaotong University · Purdue University +1
MixQuant: Adaptive Mixed-Precision Quantization for Large Language Models
Jul 25, 2026Ashitabh Misra, Madhav Agrawal, Arham Jain +1Mixed-Precision Quantization LLM Quantization
University of Illinois at Urbana-Champaign

Oct 7, 2026J/K move · Enter open · S save

Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization

Samy Houache, Yann Traonmilin, Jean-François Aujol

IMB, UB · UB, IMB

Post-Training Quantization Neural Network Quantization Mixed-Precision Quantization

Open paper AskPDF arXiv

Mixed-precision post-training quantization is a network compression method that assigns bits layer by layer, under a global memory budget using a small calibration set. The main difficulties are to overcome the combinatorial nature of the allocation problem and to manage the sensitivity to small, potentially corrupted databases. Hence, an efficient allocation method should be fast to compute and preserve model quality when calibration data are corrupted. To design such a method, we derive a layerwise probabilistic analysis of the quantization error that separates propagated error from the local perturbation introduced at a given layer. We use this local term to build a separable score for a simple allocation algorithm, that requires no external solver. The probabilistic nature of our approach brings robustness to corrupted data. On denoising tasks with DRUNet, with an average budget of 4 bits per weight, our method matches or improves state-of-the-art mixed-precision baselines under clean calibration, and is more robust to corrupted calibration, with PSNR gains of up to 7.5 dB under the tested corruptions. Experiments show bit-allocation speed-ups from 28x to 2,570x over the studied baselines. For quantized diffusion models, our experiments show that a direct application of our framework also improves the state-of-the-art.

Abstract

Explore similar work

Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization

Weight Group-wise Post-Training Quantization for Medical Foundation Model

MixQuant: Adaptive Mixed-Precision Quantization for Large Language Models