Interpretable Tensor Fusion

TL;DR

Interpretable Tensor Fusion (InTense) is a novel multimodal neural network method that learns to fuse diverse data types like text, images, and audio while providing interpretability through relevance scores, outperforming existing approaches.

Contribution

The paper introduces InTense, a theoretically grounded multimodal fusion method that captures both linear and multiplicative interactions and offers interpretability out of the box.

Findings

InTense outperforms state-of-the-art methods in accuracy.

InTense provides meaningful relevance scores on real-world datasets.

The approach effectively disentangles modality interactions.

Abstract

Conventional machine learning methods are predominantly designed to predict outcomes based on a single data type. However, practical applications may encompass data of diverse types, such as text, images, and audio. We introduce interpretable tensor fusion (InTense), a multimodal learning method for training neural networks to simultaneously learn multimodal data representations and their interpretable fusion. InTense can separately capture both linear combinations and multiplicative interactions of diverse data types, thereby disentangling higher-order interactions from the individual effects of each modality. InTense provides interpretability out of the box by assigning relevance scores to modalities and their associations. The approach is theoretically grounded and yields meaningful relevance scores on multiple synthetic and real-world datasets. Experiments on six real-world datasets show that InTense outperforms existing state-of-the-art multimodal interpretable approaches in terms of accuracy and interpretability.