Traditionally, optimizers are “programmed” to optimize queries following a set of buildin procedures. However, optimizers should be robust to its changing environment to gener...
Recent algorithmic and theoretical advances in reinforcement learning (RL) have attracted widespread interest. RL algorithmshave appeared that approximatedynamic programming on an ...
We propose a method to construct computer vision systems using a workbench composed of a multi-faceted toolbox and a general purpose kernel. The toolbox is composed of an open set ...
Essentially all data mining algorithms assume that the datagenerating process is independent of the data miner's activities. However, in many domains, including spam detectio...
Nilesh N. Dalvi, Pedro Domingos, Mausam, Sumit K. ...
In this paper we discuss a data mining framework for constructing intrusion detection models. The key ideas are to mine system audit data for consistent and useful patterns of pro...