We propose an online topic model for sequentially analyzing the time evolution of topics in document collections. Topics naturally evolve with multiple timescales. For example, so...
We analyze the regret, measured in terms of log loss, of the maximum likelihood (ML) sequential prediction strategy. This "follow the leader" strategy also defines one o...
A new algorithm for the design of complex features, to be used in the discriminant saliency approach to object classification, is presented. The algorithm consists of sequential r...
This paper extends the game-theoretic notion of internal regret to the case of on-line potfolio selection problems. New sequential investment strategies are designed to minimize th...
: A new supervised learning procedure for training RBF networks is proposed. It uses a pair of parallel running Kalman filters to sequentially update both the output weights and th...