LogART: Pushing the Limit of Efficient Logarithmic Post-Training Quantization

Jiawei Xu, Yi Zheng, Chenghe Sun, Taiyu Zhou, Zuqi Zhang, Jie Li, Lirong Zheng, Zhuo Zou

Keywords: Post-training quantization, Logarithmic quantization, Adaptive rounding, Hyperparameter search, Low-power computing

TL;DR: LogART introduces a novel logarithmic quantizer and learnable logarithmic rounding to to improve accuracy in hardware-efficient post-training quantization.

Abstract: Efficient deployment of deep neural networks increasingly relies on Post-Training Quantization (PTQ). Logarithmic PTQ, in particular, promises multiplier-free hardware efficiency, but its performance is often limited by the nonlinear and symmetric quantization grid and standard rounding-to-nearest (RTN) approach. While learnable rounding has significantly advanced linear PTQ, its application to the non-linear and often discrete nature of logarithmic domain remains unexplored. This paper introduces learnable Logarithmic Adaptive Rounding Techniques (LogART) that pioneer task-aware learnable rounding specifically for the logarithmic domain. LogART further extends the learnable rounding strategy to flexibly support outlier-aware, asymmetric, and hardware-friendly dynamic logarithmic bases, determined in a distribution-aware manner using an efficient search strategy. Extensive experiments demonstrate that LogART achieves state-of-the-art accuracy while maintaining efficiency in quantizing models across various architectures and ultra-low bitwidths, outperforming existing logarithmic PTQ methods and paving the way for more effective hardware deployment. The code is available at https://github.com/logart-lab/logart.

Primary Area: optimization

Submission Number: 1753