Grant: Research Tools and Instruments - Category 1 (<$150,000)
University of Guelph
$93,258.00
GPU infrastructure for large-scale machine learning
EF 2012-13
Résumé
Machine Learning seeks to advance the state of intelligent devices that can perform human-like reasoning: not by following rules but by learning from data. Specifically, my research group focuses on an area known as Deep Learning: methods that can automatically extract representations and abstractions from data to facilitate decision-making under uncertainty. The discovery of new algorithms and architectures requires experimentation that is computationally intensive. As humans learn from rich sensory input, our models learn from copious amounts of data that must be stored and processed.
Recent advances in parallel computing hardware, such as multicore processors, graphics processing units (GPUs), field-programmable gate arrays (FPGAs), and cloud clusters provide new opportunities for empirical investigation of machine learning. Deep Learning research is particularly well-suited to parallel hardware. The data parallelism inherent in the pixel-based inputs we often use (e.g. images and video) manifests itself in operations that apply concurrently to all pixels or local regions. In addition, task parallelism is inherent in our biologically-inspired models which are composed of redundant processing units.
GPUs are among the easiest of parallel hardware to program. They are also much more cost effective and energy efficient than CPU-based cluster computing. For these reasons, they have completely revolutionized the way our field conducts experiments. This proposal seeks a GPU computing and storage cluster: the first of its kind at the University of Guelph. This infrastructure will enable my group to conduct large-scale, world-class machine learning research that directly impacts fields as diverse as biology, entertainment, and finance. Access to this specialized parallel hardware will contribute to the training of high-quality personnel who can satiate industry's growing demand for data scientists.
Programme
Research Tools and Instruments - Category 1 (<$150,000)
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
439719-2013
Name nom
Taylor, Graham
Department département
Engineering, School of
Organization id
24
Institution établissement
University of Guelph
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2012
Competition year année de concours
2013
Award amount
93258
Program id
EQPEQ
Program name en
Research Tools and Instruments - Category 1 (<$150,000)
Program name fr
Outils et instruments de recherche - catégorie 1 (<150 000$)
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
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
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
1
Nb partie
1
Application title
GPU infrastructure for large-scale machine learning
Keyword
NSERC
Application summary
Machine Learning seeks to advance the state of intelligent devices that can perform human-like reasoning: not by following rules but by learning from data. Specifically, my research group focuses on an area known as Deep Learning: methods that can automatically extract representations and abstractions from data to facilitate decision-making under uncertainty. The discovery of new algorithms and architectures requires experimentation that is computationally intensive. As humans learn from rich sensory input, our models learn from copious amounts of data that must be stored and processed.
Recent advances in parallel computing hardware, such as multicore processors, graphics processing units (GPUs), field-programmable gate arrays (FPGAs), and cloud clusters provide new opportunities for empirical investigation of machine learning. Deep Learning research is particularly well-suited to parallel hardware. The data parallelism inherent in the pixel-based inputs we often use (e.g. images and video) manifests itself in operations that apply concurrently to all pixels or local regions. In addition, task parallelism is inherent in our biologically-inspired models which are composed of redundant processing units.
GPUs are among the easiest of parallel hardware to program. They are also much more cost effective and energy efficient than CPU-based cluster computing. For these reasons, they have completely revolutionized the way our field conducts experiments. This proposal seeks a GPU computing and storage cluster: the first of its kind at the University of Guelph. This infrastructure will enable my group to conduct large-scale, world-class machine learning research that directly impacts fields as diverse as biology, entertainment, and finance. Access to this specialized parallel hardware will contribute to the training of high-quality personnel who can satiate industry's growing demand for data scientists.