|
|
FaSS: Ensembles for Stable Learners |
|
|
This presentation introduces a new ensemble approach, Feature-Space Subdivision (FaSS), which builds local models instead of global models. FaSS is a generic ensemble approach that can use either stable or unstable models as its base models. In contrast, existing ensemble approaches which employ randomisation can only use unstable models. Our analysis shows that the new approach reduces the execution time to generate a model in an ensemble with an increased level of localisation in FaSS. Our empirical evaluation shows that FaSS performs significantly better than boosting in terms of predictive accuracy, when a stable learner SVM is used as the base learner. The speed up achieved by FaSS makes SVM ensembles a reality that would otherwise infeasible for large data sets, and FaSS SVM performs better than Boosting J48 and Random Forests when SVM is the preferred base learner. |
|
|
Jonathan Wells started out in computing in the 1980's. Initially, in game development areas such as move computations (ie. MinMax and AlphaBeta algorithms) and evaluation of positions. In the early 1990's, he started working in the area of remote controlling of scientific equipments from the PCs. He significantly updated an existing system to use what known as "client/server" model. He started his studies at Monash in 2000 and completed his Bachelor of Computing in 2004; he has just submitted his Master of Information Technology (Research) this year. Currently, he works in the Bionics and Cognitive Science Centre in the area of Haptic and Virtual Reality. |
|
