Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Waterloo

$28,000.00

Computational efficient algorithms for statistical problems

FY 2024-25

Summary
For algorithms to stay relevant and essential in our data-driven society, they must undergo continuous improvement and adaptation. Two key ways to improve algorithms involves reducing the computational time as well as the required number of iterations. This can ultimately save both time and money. Through rigorous development and accessibility, several statistical problems can be resolved with efficient algorithms to the betterment of all. The proposed research outlines the best steps to implement this improvement. First, the research will expand the efficiency of an algorithm to a much wider class of problems or models. This approach will be based on using search directions from differences in functions instead of the typical approach of differences in parameters. Second, the techniques developed in the program will aid in the identification of groups within data, or `clustering'. Advanced statistical clustering techniques are important in a variety of areas. As data `in the wild' often violate the assumption of elliptically shaped groups, much effort has been expended to provide more flexible distributional assumptions to handle increased kurtosis, skewness, and other generally non-elliptical behaviour. The applicant will investigate ways to handle data with these usual characteristics while maintaining computational efficiency. Third, contemporary research is confronted with ever-growing masses of data. Newer and faster algorithms are required to fit model and provide measures of uncertainly. This pivotal research program will explore approximation methods to estimate the fitted model and efficiency sampling methods to estimate the uncertainly of the fitted model. Top-tier students involved in the research will gain the skills essential in creating new algorithms needed to solve new problems.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2024-03994
Name nom
Browne, Ryan
Department département
Statistics and Actuarial Science
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
28000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
3001
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistical theory
Research subject fr
Théorie de statistique
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Computational efficient algorithms for statistical problems
Keywords
sketching; expectation maximization; finite mixture; block diagonal matrix; stochastic expectation maximization; measures of agreement; model based clustering
Application summary
For algorithms to stay relevant and essential in our data-driven society, they must undergo continuous improvement and adaptation. Two key ways to improve algorithms involves reducing the computational time as well as the required number of iterations. This can ultimately save both time and money. Through rigorous development and accessibility, several statistical problems can be resolved with efficient algorithms to the betterment of all. The proposed research outlines the best steps to implement this improvement. First, the research will expand the efficiency of an algorithm to a much wider class of problems or models. This approach will be based on using search directions from differences in functions instead of the typical approach of differences in parameters. Second, the techniques developed in the program will aid in the identification of groups within data, or `clustering'. Advanced statistical clustering techniques are important in a variety of areas. As data `in the wild' often violate the assumption of elliptically shaped groups, much effort has been expended to provide more flexible distributional assumptions to handle increased kurtosis, skewness, and other generally non-elliptical behaviour. The applicant will investigate ways to handle data with these usual characteristics while maintaining computational efficiency. Third, contemporary research is confronted with ever-growing masses of data. Newer and faster algorithms are required to fit model and provide measures of uncertainly. This pivotal research program will explore approximation methods to estimate the fitted model and efficiency sampling methods to estimate the uncertainly of the fitted model. Top-tier students involved in the research will gain the skills essential in creating new algorithms needed to solve new problems.