Online Passive-Aggressive Algorithms

Part of Advances in Neural Information Processing Systems 16 (NIPS 2003)

Bibtex Metadata Paper


Shai Shalev-shwartz, Koby Crammer, Ofer Dekel, Yoram Singer


We present a unified view for online classification, regression, and uni- class problems. This view leads to a single algorithmic framework for the three problems. We prove worst case loss bounds for various algorithms for both the realizable case and the non-realizable case. A conversion of our main online algorithm to the setting of batch learning is also dis- cussed. The end result is new algorithms and accompanying loss bounds for the hinge-loss.