Roland Memisevic is a computer scientist whose research interests are in machine learning and computer vision.
He develops algorithms that extract information from large amounts of data, with a particular focus on the extraction of spatial and spatio-temporal features from images and videos. Memisevic is also interested in the development of deep learning architectures and bio-inspired models, and their applications in automatic data analysis.
NSERC Discovery Grant, 2013
Memisevic, R., and G.E. Hinton. "Learning to represent spatial transformations with factored higher-order Boltzmann machines." Neural Computation 22, no. 6 (June 2010): 1473–1492.