Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications

In many machine learning systems that jointly learn from multiple modalities, a core research question is to understand the nature of multimodal interactions: how modalities combine to provide new task-relevant information that was not present in either alone. We study this challenge of interaction...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Paul Pu Liang, Chun, Kai Ling, Cheng, Yun, Obolenskiy, Alex, Liu, Yudong, Pandey, Rohan, Wilf, Alex, Louis-Philippe Morency, Salakhutdinov, Ruslan
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 13.06.2024
Subjects
Algorithms
Audio data
Couplings
Information theory
Lower bounds
Machine learning
Self-supervised learning
Upper bounds
Online AccessGet full text

Cover

More Information
Summary:In many machine learning systems that jointly learn from multiple modalities, a core research question is to understand the nature of multimodal interactions: how modalities combine to provide new task-relevant information that was not present in either alone. We study this challenge of interaction quantification in a semi-supervised setting with only labeled unimodal data and naturally co-occurring multimodal data (e.g., unlabeled images and captions, video and corresponding audio) but when labeling them is time-consuming. Using a precise information-theoretic definition of interactions, our key contribution is the derivation of lower and upper bounds to quantify the amount of multimodal interactions in this semi-supervised setting. We propose two lower bounds: one based on the shared information between modalities and the other based on disagreement between separately trained unimodal classifiers, and derive an upper bound through connections to approximate algorithms for min-entropy couplings. We validate these estimated bounds and show how they accurately track true interactions. Finally, we show how these theoretical results can be used to estimate multimodal model performance, guide data collection, and select appropriate multimodal models for various tasks.
ISSN:2331-8422