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

by   Lars Schmarje, et al.

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 consistent predictions of substructures.


page 1

page 2

page 3

page 4


Fuzzy Overclustering: Semi-Supervised Classification of Fuzzy Labels with Overclustering and Inverse Cross-Entropy

Deep learning has been successfully applied to many classification probl...

S2C2 - An orthogonal method for Semi-Supervised Learning on fuzzy labels

Semi-Supervised Learning (SSL) can decrease the amount of required label...

Life is not black and white – Combining Semi-Supervised Learning with fuzzy labels

The required amount of labeled data is one of the biggest issues in deep...

A Pitfall of Learning from User-generated Data: In-depth Analysis of Subjective Class Problem

Research in the supervised learning algorithms field implicitly assumes ...

Cost-sensitive Semi-supervised Classification for Fraud Applications

This research explores Cost-Sensitive Learning (CSL) in the fraud detect...

Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors

The performance of existing single-view 3D reconstruction methods heavil...

Please sign up or login with your details

Forgot password? Click here to reset