Saved in:
Bibliographic Details
Main Authors: Pellegrain, Victor, Tami, Myriam, Batteux, Michel, Hudelot, Céline
Format: Preprint
Published: 2021
Subjects:
Online Access:https://arxiv.org/abs/2110.08021
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911781404803072
author Pellegrain, Victor
Tami, Myriam
Batteux, Michel
Hudelot, Céline
author_facet Pellegrain, Victor
Tami, Myriam
Batteux, Michel
Hudelot, Céline
contents The increasing complexity of Industry 4.0 systems brings new challenges regarding predictive maintenance tasks such as fault detection and diagnosis. A corresponding and realistic setting includes multi-source data streams from different modalities, such as sensors measurements time series, machine images, textual maintenance reports, etc. These heterogeneous multimodal streams also differ in their acquisition frequency, may embed temporally unaligned information and can be arbitrarily long, depending on the considered system and task. Whereas multimodal fusion has been largely studied in a static setting, to the best of our knowledge, there exists no previous work considering arbitrarily long multimodal streams alongside with related tasks such as prediction across time. Thus, in this paper, we first formalize this paradigm of heterogeneous multimodal learning in a streaming setting as a new one. To tackle this challenge, we propose StreaMulT, a Streaming Multimodal Transformer relying on cross-modal attention and on a memory bank to process arbitrarily long input sequences at training time and run in a streaming way at inference. StreaMulT improves the state-of-the-art metrics on CMU-MOSEI dataset for Multimodal Sentiment Analysis task, while being able to deal with much longer inputs than other multimodal models. The conducted experiments eventually highlight the importance of the textual embedding layer, questioning recent improvements in Multimodal Sentiment Analysis benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2110_08021
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle StreaMulT: Streaming Multimodal Transformer for Heterogeneous and Arbitrary Long Sequential Data
Pellegrain, Victor
Tami, Myriam
Batteux, Michel
Hudelot, Céline
Machine Learning
Computation and Language
Multimedia
The increasing complexity of Industry 4.0 systems brings new challenges regarding predictive maintenance tasks such as fault detection and diagnosis. A corresponding and realistic setting includes multi-source data streams from different modalities, such as sensors measurements time series, machine images, textual maintenance reports, etc. These heterogeneous multimodal streams also differ in their acquisition frequency, may embed temporally unaligned information and can be arbitrarily long, depending on the considered system and task. Whereas multimodal fusion has been largely studied in a static setting, to the best of our knowledge, there exists no previous work considering arbitrarily long multimodal streams alongside with related tasks such as prediction across time. Thus, in this paper, we first formalize this paradigm of heterogeneous multimodal learning in a streaming setting as a new one. To tackle this challenge, we propose StreaMulT, a Streaming Multimodal Transformer relying on cross-modal attention and on a memory bank to process arbitrarily long input sequences at training time and run in a streaming way at inference. StreaMulT improves the state-of-the-art metrics on CMU-MOSEI dataset for Multimodal Sentiment Analysis task, while being able to deal with much longer inputs than other multimodal models. The conducted experiments eventually highlight the importance of the textual embedding layer, questioning recent improvements in Multimodal Sentiment Analysis benchmarks.
title StreaMulT: Streaming Multimodal Transformer for Heterogeneous and Arbitrary Long Sequential Data
topic Machine Learning
Computation and Language
Multimedia
url https://arxiv.org/abs/2110.08021