We develop a theory of online learning by defining several complexity measures. Among them are analogues of Rademacher complexity, covering numbers and fatshattering dimension fro...
Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
We present a unified view of two state-of-theart non-projective dependency parsers, both approximate: the loopy belief propagation parser of Smith and Eisner (2008) and the relaxe...
When applying aggregating strategies to Prediction with Expert Advice, the learning rate must be adaptively tuned. The natural choice of complexity/current loss renders the analys...
This paper presents Fuzzy-UCS, a Michigan-style Learning Fuzzy-Classifier System designed for supervised learning tasks. FuzzyUCS combines the generalization capabilities of UCS w...
Albert Orriols-Puig, Jorge Casillas, Ester Bernad&...
Most of supervised learning algorithms assume the stability of the target concept over time. Nevertheless in many real-user modeling systems, where the data is collected over an ex...