CVDec 3, 2020

Beyond Cats and Dogs: Semi-supervised Classification of fuzzy labels with overclustering

arXiv:2012.01768v20.0019 citations
AI Analysis50

This work is significant for researchers and practitioners dealing with real-world datasets where obtaining consistently labeled data is difficult due to inherent ambiguity or expert disagreement, offering an incremental improvement to semi-supervised learning.

This paper addresses the challenge of semi-supervised classification with fuzzy labels, where different experts may have varying opinions. The authors propose a novel framework utilizing overclustering to identify substructures within these fuzzy labels, demonstrating improved speed and overclustering performance on STL-10 and outperforming state-of-the-art semi-supervised methods on a plankton dataset, achieving 5-10% more consistent predictions of substructures.

A long-standing issue with deep learning is the need for large and consistently labeled datasets. Although the current research in semi-supervised learning can decrease the required amount of annotated data by a factor of 10 or even more, this line of research still uses distinct classes like cats and dogs. However, in the real-world we often encounter problems where different experts have different opinions, thus producing fuzzy labels. We propose a novel framework for handling semi-supervised classifications of such fuzzy labels. Our framework is based on the idea of overclustering to detect substructures in these fuzzy labels. We propose a novel loss to improve the overclustering capability of our framework and show on the common image classification dataset STL-10 that it is faster and has better overclustering performance than previous work. On a real-world plankton dataset, we illustrate the benefit of overclustering for fuzzy labels and show that we beat previous state-of-the-art semisupervised methods. Moreover, we acquire 5 to 10% more consistent predictions of substructures.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes