This paper describes an approach, called PRAX, to learning descriptions of a large number of texture concepts from texture samples. The learning process consists of two phases: 1) learning descriptions of selected subset of texture classes, called principal axes (briefly, praxes), and 2) learning descriptions of other classes (non-prax classes), by relating them to the praxes. Descriptions of n...

A useful way to explain the notions of implicit and explicit learning in ACT-R is to define implicit learning as learning by ACT-R's learning mechanisms, and explicit learning as the results of learning goals. This idea complies with the usual notion of implicit learning as unconscious and always active and explicit learning as intentional and conscious. Two models will be discussed to illustra...

One of the key problems in developing standard based adaptive courses is the complexity involved in the design phase, especially when establishing the hooks for the dynamic modelling to be performed at runtime. This is particularly critical when the courses are based on adaptation-oriented learning scenarios, where the full eLearning cycle (design, publication, use and auditing) is considered. ...

This work introduces an original method for registering pairs of 3D views consisting of range data sets which operates in the frequency domain. The Fourier transform allows the decoupling of the estimate of the rotation parameters from the estimate of the translation parameters, our algorithm exploits this well-known property by suggesting a three-step procedure. The rotation parameters are est...

This document provides additional details for paper Quantitative evaluation of confidence measures in a machine learning world. Single column format has been chosen to improve readability and to allow large tables to be fully displayed. Section 1 describes in detail all the confidence measures, their formulation and parameters. Section 2 reports execution time for, respectively, standalone and ...