The quest for fast learning from few examples

Though the data in our disposal are numerous and diverse, deriving meaning from them is often non trivial. This talk centers on two key challenges of data analysis, relating to the sample complexity (how many examples suffice to learn something with statistical significance) and computational complexity (how long does the computation take) of learning algorithms. In particular, we are going to consider two famous algorithms and ask what can they learn when given very few examples or a fraction of the computation time. The talk will then move on to consider why deep learning works so well for grid-structured data such as images and speech, and whether its success can be replicated for data whose inherent structure is captured by a graph. 

Andreas Loukas received a doctorate in computer science from Delft University of Technology, The Netherlands, where he focused on graph algorithms for signal processing. He is currently a research scientist at the LTS2 Signal Processing Lab in EPFL, Switzerland. His research interests lie in the intersection(s) of graph theory, high dimensional data analysis, machine learning, and signal processing.