Deep Online Learning with Stochastic Constraints

TL;DR

This paper introduces a novel online deep learning training method capable of handling multiple objectives simultaneously, demonstrating its effectiveness on Neyman-Pearson classification tasks across various benchmark datasets.

Contribution

A new online deep learning training procedure that accommodates multiple loss functions regardless of network architecture, addressing a key challenge in sequential data learning.

Findings

Effective on Neyman-Pearson classification tasks

Works across various neural network architectures

Shows improved performance on benchmark datasets

Abstract

Deep learning models are considered to be state-of-the-art in many offline machine learning tasks. However, many of the techniques developed are not suitable for online learning tasks. The problem of using deep learning models with sequential data becomes even harder when several loss functions need to be considered simultaneously, as in many real-world applications. In this paper, we, therefore, propose a novel online deep learning training procedure which can be used regardless of the neural network's architecture, aiming to deal with the multiple objectives case. We demonstrate and show the effectiveness of our algorithm on the Neyman-Pearson classification problem on several benchmark datasets.