Direction Concentration Learning: Enhancing Congruency in Machine Learning

TL;DR

This paper introduces Direction Concentration Learning (DCL), a novel method to improve congruency in machine learning, leading to better convergence and performance across various vision tasks, while also reducing catastrophic forgetting.

Contribution

DCL is a new technique that enhances congruency in learning, improving convergence paths and generalization in multiple vision tasks, and mitigating catastrophic forgetting.

Findings

DCL improves performance in saliency prediction, continual learning, and classification.

DCL reduces catastrophic forgetting in continual learning.

DCL generalizes well across models and optimizers.

Abstract

One of the well-known challenges in computer vision tasks is the visual diversity of images, which could result in an agreement or disagreement between the learned knowledge and the visual content exhibited by the current observation. In this work, we first define such an agreement in a concepts learning process as congruency. Formally, given a particular task and sufficiently large dataset, the congruency issue occurs in the learning process whereby the task-specific semantics in the training data are highly varying. We propose a Direction Concentration Learning (DCL) method to improve congruency in the learning process, where enhancing congruency influences the convergence path to be less circuitous. The experimental results show that the proposed DCL method generalizes to state-of-the-art models and optimizers, as well as improves the performances of saliency prediction task, continual learning task, and classification task. Moreover, it helps mitigate the catastrophic forgetting problem in the continual learning task. The code is publicly available at https://github.com/luoyan407/congruency.