Self-Attention Neural Bag-of-Features

TL;DR

This paper introduces novel self-attention formulations for multivariate sequence data, enhancing feature and temporal relevance modeling, and demonstrates improved performance in sequence analysis tasks using a neural bag-of-features approach.

Contribution

It proposes new self-attention mechanisms, including a joint feature-temporal attention, for better relevance quantification in sequence data analysis.

Findings

Improved accuracy over standard methods in sequence tasks

Effective joint feature-temporal attention mechanism

Versatile application with neural bag-of-features

Abstract

In this work, we propose several attention formulations for multivariate sequence data. We build on top of the recently introduced 2D-Attention and reformulate the attention learning methodology by quantifying the relevance of feature/temporal dimensions through latent spaces based on self-attention rather than learning them directly. In addition, we propose a joint feature-temporal attention mechanism that learns a joint 2D attention mask highlighting relevant information without treating feature and temporal representations independently. The proposed approaches can be used in various architectures and we specifically evaluate their application together with Neural Bag of Features feature extraction module. Experiments on several sequence data analysis tasks show the improved performance yielded by our approach compared to standard methods.