TruncQuant: Truncation-Ready Quantization for DNNs with Flexible Weight Bit Precision

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Jinhee, Yoon, Seoyeon, Lee, Taeho, Lee, Joo Chan, Jeon, Kang Eun, Ko, Jong Hwan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911003533377536
author Kim, Jinhee
Yoon, Seoyeon
Lee, Taeho
Lee, Joo Chan
Jeon, Kang Eun
Ko, Jong Hwan
author_facet Kim, Jinhee
Yoon, Seoyeon
Lee, Taeho
Lee, Joo Chan
Jeon, Kang Eun
Ko, Jong Hwan
contents The deployment of deep neural networks on edge devices is a challenging task due to the increasing complexity of state-of-the-art models, requiring efforts to reduce model size and inference latency. Recent studies explore models operating at diverse quantization settings to find the optimal point that balances computational efficiency and accuracy. Truncation, an effective approach for achieving lower bit precision mapping, enables a single model to adapt to various hardware platforms with little to no cost. However, formulating a training scheme for deep neural networks to withstand the associated errors introduced by truncation remains a challenge, as the current quantization-aware training schemes are not designed for the truncation process. We propose TruncQuant, a novel truncation-ready training scheme allowing flexible bit precision through bit-shifting in runtime. We achieve this by aligning TruncQuant with the output of the truncation process, demonstrating strong robustness across bit-width settings, and offering an easily implementable training scheme within existing quantization-aware frameworks. Our code is released at https://github.com/a2jinhee/TruncQuant.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11431
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TruncQuant: Truncation-Ready Quantization for DNNs with Flexible Weight Bit Precision
Kim, Jinhee
Yoon, Seoyeon
Lee, Taeho
Lee, Joo Chan
Jeon, Kang Eun
Ko, Jong Hwan
Machine Learning
The deployment of deep neural networks on edge devices is a challenging task due to the increasing complexity of state-of-the-art models, requiring efforts to reduce model size and inference latency. Recent studies explore models operating at diverse quantization settings to find the optimal point that balances computational efficiency and accuracy. Truncation, an effective approach for achieving lower bit precision mapping, enables a single model to adapt to various hardware platforms with little to no cost. However, formulating a training scheme for deep neural networks to withstand the associated errors introduced by truncation remains a challenge, as the current quantization-aware training schemes are not designed for the truncation process. We propose TruncQuant, a novel truncation-ready training scheme allowing flexible bit precision through bit-shifting in runtime. We achieve this by aligning TruncQuant with the output of the truncation process, demonstrating strong robustness across bit-width settings, and offering an easily implementable training scheme within existing quantization-aware frameworks. Our code is released at https://github.com/a2jinhee/TruncQuant.
title TruncQuant: Truncation-Ready Quantization for DNNs with Flexible Weight Bit Precision
topic Machine Learning
url https://arxiv.org/abs/2506.11431