Progressive Stage-wise Learning for Unsupervised Feature Representation Enhancement

TL;DR

This paper introduces Progressive Stage-wise Learning (PSL), a framework that enhances unsupervised feature representation by progressively training networks through multilevel tasks, leading to improved performance.

Contribution

The paper proposes a novel PSL framework that structures unsupervised learning into stages focusing on different task complexities, improving feature extraction.

Findings

PSL consistently improves unsupervised learning results.

Stage-wise training enhances feature representation quality.

Effective across multiple unsupervised methods.

Abstract

Unsupervised learning methods have recently shown their competitiveness against supervised training. Typically, these methods use a single objective to train the entire network. But one distinct advantage of unsupervised over supervised learning is that the former possesses more variety and freedom in designing the objective. In this work, we explore new dimensions of unsupervised learning by proposing the Progressive Stage-wise Learning (PSL) framework. For a given unsupervised task, we design multilevel tasks and define different learning stages for the deep network. Early learning stages are forced to focus on lowlevel tasks while late stages are guided to extract deeper information through harder tasks. We discover that by progressive stage-wise learning, unsupervised feature representation can be effectively enhanced. Our extensive experiments show that PSL consistently improves results for the leading unsupervised learning methods.