We examine the problem of evaluating a policy in the contextual bandit setting using only observations collected during the execution of another policy. We show that policy evalua...
John Langford, Alexander L. Strehl, Jennifer Wortm...
We relate two problems that have been explored in two distinct communities. The first is the problem of combining expert advice, studied extensively in the computational learning...
Machine learning (ML) is often used to obtain control knowledge to improve planning efficiency. Usually, ML techniques are used in isolation from experience that could be obtained...
In this paper we show how frequent sequence mining (FSM) can be applied to data produced by monitoring distributed enterprise applications. In particular we show how we applied FSM...
The design of complex systems, e.g., telecom services, is nowadays usually based on the integration of components (COTS), loosely coupled in distributed architectures. When compon...