In multi-task learning several related tasks are considered simultaneously, with the hope that by an appropriate sharing of information across tasks, each task may benefit from th...
We propose an approach that takes as input a task model, which includes the user's view of the interactive system, and automatically discovers a set of categorized and ranked...
Abstract. Active learning refers to the task of devising a ranking function that, given a classifier trained from relatively few training examples, ranks a set of additional unlabe...
We present a new family of subgradient methods that dynamically incorporate knowledge of the geometry of the data observed in earlier iterations to perform more informative gradie...
Abstract. This work proposes a family of language-independent semantic kernel functions defined for individuals in an ontology. This allows exploiting wellfounded kernel methods fo...