Deep Learning Ensembles in Toupee

Abstract: In Deep Learning, Ensembles are usually created as an afterthought. Using Toupee, the process of creating an Ensemble becomes integral to the Deep Learning process, with novel ensemble methods that integrate with the deep learning algorithms at hand. I will show how toupee works and explore some of these methods, using standard computer vision benchmark datasets.

Bio: Alan Mosca is a doctoral researcher at Birkbeck, University of London, where his research focuses on deep learning ensembles and improvements to optimization algorithms in deep learning. Previously, Alan worked at Wadhwani Asset Management, Jane Street Capital, and several software companies as well as on many consulting projects in machine learning and deep learning.