paper-with-me

Papers

PACT: Parameterized Clipping Activation for Quantized Neural Networks

2018-05-16 · ICLR 2018 1 · Jungwook Choi, Zhuo Wang, Swagath Venkataramani, Pierce I-Jen Chuang, Vijayalakshmi Srinivasan, Kailash Gopalakrishnan

Deep learning algorithms achieve high classification accuracy at the expense of significant computation cost. To address this cost, a number of quantization schemes have been proposed - but most of these techniques focused on quantizing weights, which are relatively smaller in size compared to activations. This paper proposes a novel quantization scheme for activations during training - that enables neural networks to work well with ultra low precision weights and activations without any significant accuracy degradation. This technique, PArameterized Clipping acTivation (PACT), uses an activation clipping parameter $\alpha$ that is optimized during training to find the right quantization scale. PACT allows quantizing activations to arbitrary bit precisions, while achieving much better accuracy relative to published state-of-the-art quantization schemes. We show, for the first time, that both weights and activations can be quantized to 4-bits of precision while still achieving accuracy comparable to full precision networks across a range of popular models and datasets. We also show that exploiting these reduced-precision computational units in hardware can enable a super-linear improvement in inferencing performance due to a significant reduction in the area of accelerator compute engines coupled with the ability to retain the quantized model and activation data in on-chip memories.

📄 PDF Abstract BibTeX arXiv:1805.06085

Code (3)

KwangHoonAn/PACT pytorch
PaddlePaddle/PaddleOCR paddle
cornell-zhang/dnn-gating pytorch

Tasks

Quantization

Similar Papers 제목 키워드 기반

Bridging the Accuracy Gap for 2-bit Quantized Neural Networks (QNN)

2018-07-17 · Jungwook Choi, Pierce I-Jen Chuang, Zhuo Wang, Swagath Venkataramani 외

Deep learning algorithms achieve high classification accuracy at the expense of significant computation cost. In order to reduce this cost, several quantization schemes have gained attention recently with some focusing o…

General ClassificationQuantization

Rethinking Neural Network Quantization

2019-09-25 · Qing Jin, Linjie Yang, Zhenyu Liao

Quantization reduces computation costs of neural networks but suffers from performance degeneration. Is this accuracy drop due to the reduced capacity, or inefficient training during the quantization procedure? After loo…

Quantization

Towards Efficient Training for Neural Network Quantization

2019-12-21 · Qing Jin, Linjie Yang, Zhenyu Liao

Quantization reduces computation costs of neural networks but suffers from performance degeneration. Is this accuracy drop due to the reduced capacity, or inefficient training during the quantization procedure? After loo…

Quantization

Quantized and Interpretable Learning Scheme for Deep Neural Networks in Classification Task

2024-12-05 · Alireza Maleki, Mahsa Lavaei, Mohsen Bagheritabar, Salar Beigzad 외

Deep learning techniques have proven highly effective in image classification, but their deployment in resourceconstrained environments remains challenging due to high computational demands. Furthermore, their interpreta…

image-classificationImage ClassificationQuantization

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

2022-02-10 · ICLR 2022 4 · Qing Jin, Jian Ren, Richard Zhuang, Sumant Hanumante 외

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 a…

Quantization