MMBind: Unleashing the Potential of Distributed and Heterogeneous Data for Multimodal Learning in IoT

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ouyang, Xiaomin, Wu, Jason, Kimura, Tomoyoshi, Lin, Yihan, Verma, Gunjan, Abdelzaher, Tarek, Srivastava, Mani
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929742322597888
author Ouyang, Xiaomin
Wu, Jason
Kimura, Tomoyoshi
Lin, Yihan
Verma, Gunjan
Abdelzaher, Tarek
Srivastava, Mani
author_facet Ouyang, Xiaomin
Wu, Jason
Kimura, Tomoyoshi
Lin, Yihan
Verma, Gunjan
Abdelzaher, Tarek
Srivastava, Mani
contents Multimodal sensing systems are increasingly prevalent in various real-world applications. Most existing multimodal learning approaches heavily rely on training with a large amount of synchronized, complete multimodal data. However, such a setting is impractical in real-world IoT sensing applications where data is typically collected by distributed nodes with heterogeneous data modalities, and is also rarely labeled. In this paper, we propose MMBind, a new data binding approach for multimodal learning on distributed and heterogeneous IoT data. The key idea of MMBind is to construct a pseudo-paired multimodal dataset for model training by binding data from disparate sources and incomplete modalities through a sufficiently descriptive shared modality. We also propose a weighted contrastive learning approach to handle domain shifts among disparate data, coupled with an adaptive multimodal learning architecture capable of training models with heterogeneous modality combinations. Evaluations on ten real-world multimodal datasets highlight that MMBind outperforms state-of-the-art baselines under varying degrees of data incompleteness and domain shift, and holds promise for advancing multimodal foundation model training in IoT applications\footnote (The source code is available via https://github.com/nesl/multimodal-bind).
format Preprint
id arxiv_https___arxiv_org_abs_2411_12126
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MMBind: Unleashing the Potential of Distributed and Heterogeneous Data for Multimodal Learning in IoT
Ouyang, Xiaomin
Wu, Jason
Kimura, Tomoyoshi
Lin, Yihan
Verma, Gunjan
Abdelzaher, Tarek
Srivastava, Mani
Machine Learning
Multimodal sensing systems are increasingly prevalent in various real-world applications. Most existing multimodal learning approaches heavily rely on training with a large amount of synchronized, complete multimodal data. However, such a setting is impractical in real-world IoT sensing applications where data is typically collected by distributed nodes with heterogeneous data modalities, and is also rarely labeled. In this paper, we propose MMBind, a new data binding approach for multimodal learning on distributed and heterogeneous IoT data. The key idea of MMBind is to construct a pseudo-paired multimodal dataset for model training by binding data from disparate sources and incomplete modalities through a sufficiently descriptive shared modality. We also propose a weighted contrastive learning approach to handle domain shifts among disparate data, coupled with an adaptive multimodal learning architecture capable of training models with heterogeneous modality combinations. Evaluations on ten real-world multimodal datasets highlight that MMBind outperforms state-of-the-art baselines under varying degrees of data incompleteness and domain shift, and holds promise for advancing multimodal foundation model training in IoT applications\footnote (The source code is available via https://github.com/nesl/multimodal-bind).
title MMBind: Unleashing the Potential of Distributed and Heterogeneous Data for Multimodal Learning in IoT
topic Machine Learning
url https://arxiv.org/abs/2411.12126