Open-World Class Discovery with Kernel Networks

Abstract

We study an Open-World Class Discovery problem in which, given labeled training samples from old classes, we need to discover new classes from unlabeled test samples. There are two critical challenges to addressing this paradigm: (a) transferring knowledge from old to new classes, and (b) incorporating knowledge learned from new classes back to the original model. We propose Class Discovery Kernel Network with Expansion (CD-KNet-Exp), a deep learning framework, which utilizes the Hilbert Schmidt Independence Criterion to bridge supervised and unsupervised information together in a systematic way, such that the learned knowledge from old classes is distilled appropriately for discovering new classes. Compared to competing methods, CD-KNet-Exp shows superior performance on three publicly available benchmark datasets and a challenging real-world radio frequency fingerprinting dataset.

arXiv commentAccepted to the IEEE International Conference on Data Mining 2020 (ICDM'20); Best paper candidate

References in corpus (7)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Semi-Supervised Learning with Ladder Networks
• Anomaly Detection using One-Class Neural Networks
• Unseen Class Discovery in Open-world Classification
• Iterative Spectral Method for Alternative Clustering
• Semi-Supervised Class Discovery