Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Toronto

$122,000.00

Machine Learning for Perception

FY 2013-14

Summary
Computers are still much worse than people at recognizing objects in images or words in sound-waves. New methods for training artificial neural networks have improved their abilities to learn multiple layers of features and this has led to significant improvements in recognition, but artificial neural networks still lag far behind the real thing. The representational power of neural networks can be greatly improved by grouping neurons into ""capsules"" that each perform a lot of internal computation and output a small set of real numbers that are bound together as properties of the same entity. By using pairs of images related by a known transformation it is possible to allow each capsule to decide for itself what kind of visual or acoustic entity it should represent but to force each capsule to output the same kinds of coordinates as are used to handle viewpoint effects in computer graphics. This allows spatial relationships to be modeled as linear operations which makes it easy to recognize objects or words by using the relationships between their parts. Low level capsules learn to represent simple entities in small regions of the image. Higher-level capsules learn to represent more complex entities and they operate over larger regions of the image but they retain precise information about the position, orientation and scale of the visual entity they represent. This allows spatial relationships between parts to be used to recognize entities at many different levels in the whole-part hierarchy.Another way to significantly improve the performance of artificial neural networks is to use a much better optimizer that was recently developed at the University of Toronto for learning the weights of a recurrent neural network. The new optimizer makes it possible to train neural networks to use regularities at many different time scales to model very complicated sequential data such as speech. We plan to combine all of the main stages of an automatic speech recognition system into a single network that can be be trained to convert sound waves into sequences of recognized words.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
9185-2012
Name nom
Hinton, Geoffrey
Department département
Computer Science (St. George Campus)
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2013
Competition year année de concours
2012
Award amount
122000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
Research subject code
2805
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Learning and inference theories
Research subject fr
Théories de l'apprentissage et l'inférence
Partie
2
Nb partie
5
Application title
Machine Learning for Perception
Keyword
NSERC
Application summary
Computers are still much worse than people at recognizing objects in images or words in sound-waves. New methods for training artificial neural networks have improved their abilities to learn multiple layers of features and this has led to significant improvements in recognition, but artificial neural networks still lag far behind the real thing. The representational power of neural networks can be greatly improved by grouping neurons into ""capsules"" that each perform a lot of internal computation and output a small set of real numbers that are bound together as properties of the same entity. By using pairs of images related by a known transformation it is possible to allow each capsule to decide for itself what kind of visual or acoustic entity it should represent but to force each capsule to output the same kinds of coordinates as are used to handle viewpoint effects in computer graphics. This allows spatial relationships to be modeled as linear operations which makes it easy to recognize objects or words by using the relationships between their parts. Low level capsules learn to represent simple entities in small regions of the image. Higher-level capsules learn to represent more complex entities and they operate over larger regions of the image but they retain precise information about the position, orientation and scale of the visual entity they represent. This allows spatial relationships between parts to be used to recognize entities at many different levels in the whole-part hierarchy.Another way to significantly improve the performance of artificial neural networks is to use a much better optimizer that was recently developed at the University of Toronto for learning the weights of a recurrent neural network. The new optimizer makes it possible to train neural networks to use regularities at many different time scales to model very complicated sequential data such as speech. We plan to combine all of the main stages of an automatic speech recognition system into a single network that can be be trained to convert sound waves into sequences of recognized words.