SLDS2015 Logo

Special Session on Prediction with Expert Advice

Chair: Yuri Kalnishkan

Prediction with expert advice is an area of machine learning concerned with merging predictions of learners, usually in the on-line mode of learning, and obtaining performance guarantees. Applications of prediction withe expert advice and related areas include on-line convex optimization, construction of universal algorithms and portfolios, which compete with parameterized classes of prediction and investment algorithms, predictive complexity, which quantifies the predictability of sequences, and other topics. Submissions are solicited on theoretical aspects of prediction with expert advice and related areas and applications to real-world data.