paper-with-me

Papers

F8Net: Fixed-Point 8-bit Only Multiplication for Network Quantization

2022-02-10 · ICLR 2022 4 · Qing Jin, Jian Ren, Richard Zhuang, Sumant Hanumante, Zhengang Li, Zhiyu Chen, Yanzhi Wang, Kaiyuan Yang, Sergey Tulyakov

Neural network quantization is a promising compression technique to reduce memory footprint and save energy consumption, potentially leading to real-time inference. However, there is a performance gap between quantized and full-precision models. To reduce it, existing quantization approaches require high-precision INT32 or full-precision multiplication during inference for scaling or dequantization. This introduces a noticeable cost in terms of memory, speed, and required energy. To tackle these issues, we present F8Net, a novel quantization framework consisting of only fixed-point 8-bit multiplication. To derive our method, we first discuss the advantages of fixed-point multiplication with different formats of fixed-point numbers and study the statistical behavior of the associated fixed-point numbers. Second, based on the statistical and algorithmic analysis, we apply different fixed-point formats for weights and activations of different layers. We introduce a novel algorithm to automatically determine the right format for each layer during training. Third, we analyze a previous quantization algorithm -- parameterized clipping activation (PACT) -- and reformulate it using fixed-point arithmetic. Finally, we unify the recently proposed method for quantization fine-tuning and our fixed-point approach to show the potential of our method. We verify F8Net on ImageNet for MobileNet V1/V2 and ResNet18/50. Our approach achieves comparable and better performance, when compared not only to existing quantization techniques with INT32 multiplication or floating-point arithmetic, but also to the full-precision counterparts, achieving state-of-the-art performance.

📄 PDF Abstract BibTeX arXiv:2202.05239

Code (1)

snap-research/f8net 공식 구현 pytorch

Tasks

Quantization

Similar Papers 제목 키워드 기반

Trainable Fixed-Point Quantization for Deep Learning Acceleration on FPGAs

2024-01-31 · Dingyi Dai, Yichi Zhang, Jiahao Zhang, Zhanqiu Hu 외

Quantization is a crucial technique for deploying deep learning models on resource-constrained devices, such as embedded FPGAs. Prior efforts mostly focus on quantizing matrix multiplications, leaving other layers like B…

Deep LearningQuantization

Accelerating Neural Network Inference by Overflow Aware Quantization

2020-05-27 · Hongwei Xie, Shuo Zhang, Huanghao Ding, Yafei Song 외

The inherent heavy computation of deep neural networks prevents their widespread applications. A widely used method for accelerating model inference is quantization, by replacing the input operands of a network using fix…

image-classificationImage Classificationobject-detectionObject Detection+2

High-Rate Quantized Matrix Multiplication I

2026-01-23 · Or Ordentlich, Yury Polyanskiy arxiv

This paper investigates the problem of quantized matrix multiplication (MatMul), which has become crucial for the efficient deployment of large language models (LLMs). We consider a Generic MatMul setting, where both mat…

BiQGEMM: Matrix Multiplication with Lookup Table For Binary-Coding-based Quantized DNNs

2020-05-20 · Yongkweon Jeon, Baeseong Park, Se Jung Kwon, Byeongwook Kim 외

The number of parameters in deep neural networks (DNNs) is rapidly increasing to support complicated tasks and to improve model accuracy. Correspondingly, the amount of computations and required memory footprint increase…

Quantization

Mix and Match: A Novel FPGA-Centric Deep Neural Network Quantization Framework

2020-12-08 · Sung-En Chang, Yanyu Li, Mengshu Sun, Runbin Shi 외

Deep Neural Networks (DNNs) have achieved extraordinary performance in various application domains. To support diverse DNN models, efficient implementations of DNN inference on edge-computing platforms, e.g., ASICs, FPGA…

Edge-computingModel CompressionQuantization