Yonatan Amit, Shai Shalev-shwartz, Yoram Singer
We describe and analyze an algorithmic framework for online classiﬁcation where each online trial consists of multiple prediction tasks that are tied together. We tackle the problem of updating the online hypothesis by deﬁning a projection problem in which each prediction task corresponds to a single linear constraint. These constraints are tied together through a single slack parameter. We then in- troduce a general method for approximately solving the problem by projecting simultaneously and independently on each constraint which corresponds to a pre- diction sub-problem, and then averaging the individual solutions. We show that this approach constitutes a feasible, albeit not necessarily optimal, solution for the original projection problem. We derive concrete simultaneous projection schemes and analyze them in the mistake bound model. We demonstrate the power of the proposed algorithm in experiments with online multiclass text categorization. Our experiments indicate that a combination of class-dependent features with the simultaneous projection method outperforms previously studied algorithms.