Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications

TL;DR

This paper introduces information-theoretic bounds to quantify multimodal interactions in semi-supervised learning, enabling better understanding and application of multimodal models without extensive labeled data.

Contribution

It provides the first theoretical bounds on multimodal interactions in semi-supervised settings, validated through empirical experiments.

Findings

Proposed bounds accurately estimate true multimodal interactions.

Bounds can guide data collection and model selection.

Validated bounds improve understanding of multimodal model performance.

Abstract

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.