paper-with-me

홈 › Papers

Value-aware Quantization for Training and Inference of Neural Networks

2018-04-20 · ECCV 2018 9 · Eunhyeok Park, Sungjoo Yoo, Peter Vajda

We propose a novel value-aware quantization which applies aggressively reduced precision to the majority of data while separately handling a small amount of large data in high precision, which reduces total quantization errors under very low precision. We present new techniques to apply the proposed quantization to training and inference. The experiments show that our method with 3-bit activations (with 2% of large ones) can give the same training accuracy as full-precision one while offering significant (41.6% and 53.7%) reductions in the memory cost of activations in ResNet-152 and Inception-v3 compared with the state-of-the-art method. Our experiments also show that deep networks such as Inception-v3, ResNet-101 and DenseNet-121 can be quantized for inference with 4-bit weights and activations (with 1% 16-bit data) within 1% top-1 accuracy drop.

📄 PDF Abstract BibTeX arXiv:1804.07802

Code (0)

등록된 구현이 없습니다.

Tasks

Quantization

Methods 이 논문이 사용한 방법론

RMSProp RMSProp is an unpublished adaptive learning rate optimizer proposed by Geoff Hinton. The motivation…
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
Average Pooling 설명 없음
Auxiliary Classifier Auxiliary Classifiers are type of architectural component that seek to improve the convergence of very deep networks. They are classifier heads we attach to layers before the…
1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…
Inception-v3 Module Inception-v3 Module is an image block used in the Inception-v3 architecture. This architecture is used on the coarsest (8 ×…
Max Pooling Max Pooling is a pooling operation that calculates the maximum value for patches of a feature map, and uses it to create a downsampled (pooled) feature map. It is usually…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

On the efficient representation and execution of deep acoustic models

2016-07-15 · Raziel Alvarez, Rohit Prabhavalkar, Anton Bakhtin

In this paper we present a simple and computationally efficient quantization scheme that enables us to reduce the resolution of the parameters of a neural network from 32-bit floating point values to 8-bit integer values…

Quantizationspeech-recognitionSpeech Recognition

Attention Round for Post-Training Quantization

2022-07-07 · Huabin Diao, Gongyan Li, Shaoyun Xu, Yuexing Hao

At present, the quantification methods of neural network models are mainly divided into post-training quantization (PTQ) and quantization aware training (QAT). Post-training quantization only need a small part of the dat…

Combinatorial OptimizationQuantization

Fairy2i: Training Complex LLMs from Real LLMs with All Parameters in $\{\pm 1, \pm i\}$

2025-12-02 · Feiyu Wang, Xinyu Tan, Bokai Huang, Yihao Zhang 외 arxiv

Large language models (LLMs) have revolutionized artificial intelligence, yet their massive memory and computational demands necessitate aggressive quantization, increasingly pushing representations toward the theoretica…

CAT: Compression-Aware Training for bandwidth reduction

2019-09-25 · Chaim Baskin, Brian Chmiel, Evgenii Zheltonozhskii, Ron Banner 외

Convolutional neural networks (CNNs) have become the dominant neural network architecture for solving visual processing tasks. One of the major obstacles hindering the ubiquitous use of CNNs for inference is their relati…

Quantization

AdaQAT: Adaptive Bit-Width Quantization-Aware Training

2024-04-22 · Cédric Gernigon, Silviu-Ioan Filip, Olivier Sentieys, Clément Coggiola 외

Large-scale deep neural networks (DNNs) have achieved remarkable success in many application scenarios. However, high computational complexity and energy costs of modern DNNs make their deployment on edge devices challen…

Quantization