October 29, 2014
A common problem in data analysis – including machine learning and genomics – is to detect, within a large array, small submatrices which are ‘structured’ in some way. Such submatrices, called ‘biclusters’ can represent a subset of features shared across a subset of images, or a subsets of genes that are coexpressed across a subset of the patient population. In this talk I will discuss some of the challenges associated with biclustering, and provide an algorithm that overcomes most of these challenges.
Dr. Aaditya V Rangan is an Assistant Professor at the Courant Institute of Mathematical Sciences, NYU. He received his PhD in Mathematics from the University of California, Berkeley. His research interests lie in large – scale scientific modeling of physical, biological and neurobiological phenomena and the development of efficient numerical methods and related analysis.