Alex Graves
Program
Junior Fellow Academy
Appointment
Junior Fellow, Neural Computation and Adaptive Perception
Institution
University of Toronto
Country
Canada 
Alex Graves is a CIFAR Junior Fellow working under the supervision of Neural Computation and Adaptive Perception (NCAP) Fellow and Program Director, Geoffrey Hinton in the Department of Computer Science, University of Toronto. He received his Ph.D. in 2008 from the Dalle Molle Institute for Artificial Intelligence, Lugano, Switzerland and the Technical University of Munich, Germany with Summa Cum Laude. His doctoral thesis advisor was Prof. Jürgen Schmidhuber. Alex also holds a Certificate of Advanced Study in Mathematics, with distinction, from the University of Cambridge and a B.Sc. with first class Honours in Mathematical Physics from the University of Edinburgh. Alex is currently completing his appointment as a postdoctoral researcher with the Chair of Robotics and Embedded Systems in the Technical University of Munich. Outside of his academic work, he also does freelance consulting and software development for handwriting recognition technology.
Alex’s research focuses on using computers to find patterns in sequential data. He is particularly interested in the patterns perceived by humans in sensory streams such as sound, vision and touch. His main computational tools are recurrent neural networks – artificial sequence processing devices inspired by the cyclical connectivity of neurons in the brain. He has previously applied such networks to classification and transcription problems such as speech and handwriting recognition. He has been notably successful in the latter area, winning cursive handwriting recognition competitions in several languages. As a Junior Fellow, Alex will concentrate on methods for discovering temporal patterns in perceptual data using information-theoretic principles alone. Such methods have a wide range of potential applications, including data compression, sequential clustering, time-series prediction and sequence labeling.
Alex’s research focuses on using computers to find patterns in sequential data. He is particularly interested in the patterns perceived by humans in sensory streams such as sound, vision and touch. His main computational tools are recurrent neural networks – artificial sequence processing devices inspired by the cyclical connectivity of neurons in the brain. He has previously applied such networks to classification and transcription problems such as speech and handwriting recognition. He has been notably successful in the latter area, winning cursive handwriting recognition competitions in several languages. As a Junior Fellow, Alex will concentrate on methods for discovering temporal patterns in perceptual data using information-theoretic principles alone. Such methods have a wide range of potential applications, including data compression, sequential clustering, time-series prediction and sequence labeling.