Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks

TL;DR

This paper introduces Self-Supervised Autogenous Learning (SSAL) models that incorporate auxiliary signals into deep neural networks, enhancing interpretability, convergence speed, and performance over state-of-the-art methods.

Contribution

The paper proposes SSAL models that integrate self-supervised auxiliary tasks into DNNs, improving interpretability and training efficiency compared to traditional single-objective approaches.

Findings

SSAL models outperform state-of-the-art classifiers.

SSAL provides more interpretable structured predictions.

SSAL models converge faster during training.

Abstract

Classification problems solved with deep neural networks (DNNs) typically rely on a closed world paradigm, and optimize over a single objective (e.g., minimization of the cross-entropy loss). This setup dismisses all kinds of supporting signals that can be used to reinforce the existence or absence of a particular pattern. The increasing need for models that are interpretable by design makes the inclusion of said contextual signals a crucial necessity. To this end, we introduce the notion of Self-Supervised Autogenous Learning (SSAL) models. A SSAL objective is realized through one or more additional targets that are derived from the original supervised classification task, following architectural principles found in multi-task learning. SSAL branches impose low-level priors into the optimization process (e.g., grouping). The ability of using SSAL branches during inference, allow models to converge faster, focusing on a richer set of class-relevant features. We show that SSAL models consistently outperform the state-of-the-art while also providing structured predictions that are more interpretable.