Preventing Local Pitfalls in Vector Quantization via Optimal Transport

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Borui, Zheng, Wenzhao, Zhou, Jie, Lu, Jiwen
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913620034584576
author Zhang, Borui
Zheng, Wenzhao
Zhou, Jie
Lu, Jiwen
author_facet Zhang, Borui
Zheng, Wenzhao
Zhou, Jie
Lu, Jiwen
contents Vector-quantized networks (VQNs) have exhibited remarkable performance across various tasks, yet they are prone to training instability, which complicates the training process due to the necessity for techniques such as subtle initialization and model distillation. In this study, we identify the local minima issue as the primary cause of this instability. To address this, we integrate an optimal transport method in place of the nearest neighbor search to achieve a more globally informed assignment. We introduce OptVQ, a novel vector quantization method that employs the Sinkhorn algorithm to optimize the optimal transport problem, thereby enhancing the stability and efficiency of the training process. To mitigate the influence of diverse data distributions on the Sinkhorn algorithm, we implement a straightforward yet effective normalization strategy. Our comprehensive experiments on image reconstruction tasks demonstrate that OptVQ achieves 100% codebook utilization and surpasses current state-of-the-art VQNs in reconstruction quality.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15195
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Preventing Local Pitfalls in Vector Quantization via Optimal Transport
Zhang, Borui
Zheng, Wenzhao
Zhou, Jie
Lu, Jiwen
Computer Vision and Pattern Recognition
Machine Learning
Vector-quantized networks (VQNs) have exhibited remarkable performance across various tasks, yet they are prone to training instability, which complicates the training process due to the necessity for techniques such as subtle initialization and model distillation. In this study, we identify the local minima issue as the primary cause of this instability. To address this, we integrate an optimal transport method in place of the nearest neighbor search to achieve a more globally informed assignment. We introduce OptVQ, a novel vector quantization method that employs the Sinkhorn algorithm to optimize the optimal transport problem, thereby enhancing the stability and efficiency of the training process. To mitigate the influence of diverse data distributions on the Sinkhorn algorithm, we implement a straightforward yet effective normalization strategy. Our comprehensive experiments on image reconstruction tasks demonstrate that OptVQ achieves 100% codebook utilization and surpasses current state-of-the-art VQNs in reconstruction quality.
title Preventing Local Pitfalls in Vector Quantization via Optimal Transport
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2412.15195