The present paper considers the effects of introducing inaccuracies in a learner’s environment in Gold’s learning model of identification in the limit. Three kinds of inaccu...
Abstract. We present a machine learning approach called shape regression machine (SRM) to segmenting in real time an anatomic structure that manifests a deformable shape in a medic...
Improving the sample efficiency of reinforcement learning algorithms to scale up to larger and more realistic domains is a current research challenge in machine learning. Model-ba...
A pattern is a finite string of constant and variable symbols. The erasing language generated by a pattern p is the set of all strings that can be obtained by substituting (possib...
The supremacy of n-gram models in statistical language modelling has recently been challenged by parametric models that use distributed representations to counteract the difficult...