Feature Mining: A Novel Training Strategy for Convolutional Neural Network

TL;DR

This paper introduces Feature Mining, a parameter-free, plug-and-play training strategy for CNNs that enhances local feature learning by dividing and reusing feature parts, improving model performance.

Contribution

The paper proposes a novel, versatile training method called Feature Mining that strengthens local feature learning in CNNs through feature segmentation and reusing, applicable to any CNN model.

Findings

Improves local feature learning in CNNs.

Compatible with various CNN architectures.

Enhances model performance across tasks.

Abstract

In this paper, we propose a novel training strategy for convolutional neural network(CNN) named Feature Mining, that aims to strengthen the network's learning of the local feature. Through experiments, we find that semantic contained in different parts of the feature is different, while the network will inevitably lose the local information during feedforward propagation. In order to enhance the learning of local feature, Feature Mining divides the complete feature into two complementary parts and reuse these divided feature to make the network learn more local information, we call the two steps as feature segmentation and feature reusing. Feature Mining is a parameter-free method and has plug-and-play nature, and can be applied to any CNN models. Extensive experiments demonstrate the wide applicability, versatility, and compatibility of our method.