Articles from:December 2024

  • Mo Ahsan Ahmad Masters Project Presentation Friday, December 20, 10:30 AM

    Tuesday, December 17, 2024 | By

    Mo Ahsan Ahmad, a Master of Science candidate in the Department of Mathematics and Statistics, will present their Masters project titled Advancing Generative Modeling and Applications with Boltzmann Machines, Restricted Boltzmann Machines, and Sum-Product Networks on Friday, December 20, 2024 at 10:30 AM online on Microsoft Teams.

    The examination committee includes supervisors Dr. Ejaz Ahmed and Dr. Yifeng Li and Supervisory Committee Member Dr. Pouria Ramazi.

    Students (both graduate and undergraduate) as well as other members of the Brock Community are invited to attend. If you are interested in the presentation, please contact Neil Marshall at [email protected] for a link to the team. Please join with your microphones and camera turned off.

    Keywords: Probabilistic Models, Boltzmann Machines, Restricted Boltzmann Machines, SPNs, FMNIST Dataset, Model Performance

    Abstract:
    We live in the era of advanced machine learning methodologies with promising applications in generative probabilistic modeling. This study explores advanced machine learning methodologies with promising applications in probabilistic modeling and real-world problem-solving. The investigation focuses on Boltzmann Machines (BMs), Restricted Boltzmann Machines (RBMs), and Sum-Product Networks (SPNs), emphasizing their ability to analyze complicated data distributions and reconstruct meaningful outcomes. BMs and RBMs, as energy-based probabilistic models, provide a strong foundation for capturing patterns in a variety of datasets. SPNs, with their hierarchical structure, allow for scalable probabilistic inference and efficient data representation. Using the Fashion MNIST dataset as a benchmark, this work demonstrates the practical performance of these models, highlighting reconstructed images, and precise predictions, alongside quantitative performance metrics. These findings are relevant for a variety of applications, such as image synthesis, object detection, and pattern recognition in domains like healthcare diagnostics and scientific research. The results highlight the distinct strengths of each strategy: the scalability and inference effectiveness of SPNs, as well as the capacity of BMs and RBMs to efficiently reconstruct and model data distributions. By providing a comparative analysis of these approaches, the study provides practical insights for both researchers and developers, demonstrating generative models’ revolutionary capability in developing machine learning and deep learning applications.

  • Congratulations to Dorothy Levay winner of the 2024 Brock President’s Distinguished Staff Award

    Friday, December 13, 2024 | By

    The Department of Mathematics and Statistics would like to congratulate our staff member, Dorothy Levay who is the recipient of the 2024 Brock President’s Distinguished Staff Award. This award is to recognize staff who, in addition to their normal duties:

    • Demonstrated exemplary service and/or make a significant contribution in his/her unit

    • Done something exceptional to advance Brock’s reputation

    • Made a significant contribution to the University and/or community

    • Provided a valuable service to the broader community at Brock University outside his/her own unit 

    Photo of Dorothy Levay and her award

    Dorothy has for many years contributed to the success of the Department of Mathematics and Statistics and Brock University. She voluntarily took on a leadership role for the Department of Mathematics and Statistics’ teaching assistants years before it was officially added to her job description. She mentors the Department’s part-time instructors and teaching assistants, including observing lectures for first-time instructors and serving as chair of the Department’s Sessional and ILTA Teaching committee. As part of this, Dorothy reviews applications, makes recommendations to other committee members, conducts interviews and provides workplace references to part-time instructors and teaching assistants as requested.

    She also steps up when needed to assist with the the Department’s teaching, including rearranging her teaching deployment on short notice when, for example, sudden enrollment required a second section of Math 1P06 (Calculus II for Scientists). While regular teaching is a part of her job description, Dorothy goes above and beyond to provide exceptional instruction to our students, as recognized by her recipient of the 2015 FMS Distinguished Teaching Award as well as the 2001 Making a Difference Award, by the Brock Centre for Students with disABILITIES (now called Student Accessibility Services)

    Dorothy has contributed to advancing Brock’s reputation, by being the principal organizer of the tribute event in recognition of Professor Emeritus Eric Muller on October 3rd, 2024, officially announcing the newly renamed ‘Eric Muller Math & Stats Learning Centre.’ She also served as the co-chair of the local organizing committee of a 4-day Professional Development and Symposium, titled Coding, Computational Modeling, & Equity in Mathematics Education from April 26-29th 2023.

    She has served the university and the community by participating on the external Best Practices Recognition Advisory committee from 2001 to 2010. Dorothy also coordinated from 1996 to 2000 the First Nations Science Camp  during summers.

    She has been involved for many years with the Department of Mathematics and Statistics’ continuous development of its large service courses Math 1P97 and Stat 1P98, which serve close to 3000 students annually. As part of this, she volunteered to join the team responsible for updating the online course modules for both courses. Dorothy took the initiative to consult client departments such as Business, Health Sciences, Biology, Computer Science and Economics) to create program-focused examples as way to make the course contents explicitly more relevant to students. She has also lead the Department’s Service Course Committee when it redeveloped the assignments for Stat 1P98 to include a group case study including real-world data. She has been involved in the creation of the original online modules for the Math 1P97 online course. This commitment to these courses dates back to much earlier times, when she supported the (at the time, innovative) integration of Minitab in STAT 1P98 and of Maple in MATH 1P97.

    We are so happy that Dorothy’s exceptional contributions to the student experience and the Brock Community have been recognized by President Lesley Rigg and the University. Congratulations again Dorothy!

  • Azar Taheri Tayebi Masters Thesis Defence Wednesday, December 18, 2:00 PM

    Thursday, December 12, 2024 | By

    Azar Taheri Tayebi, a Master of Science candidate in the Department of Mathematics and Statistics, will present their Masters thesis titled Use of citizen-reports on angler behavior on Wednesday, December 18, 2024 at 2:00 PM in-person on campus.

    The examination committee includes Stephen Anco, Chair; Pouria Ramazi, Supervisor; Brett van Poorten, External Examiner (Simon Fraser University); and Yifeng Li and William Marshall, Supervisory Committee Members.

    Students (both graduate and undergraduate) as well as other members of the Brock Community are invited to attend. If you are interested in the presentation, please contact Neil Marshall at [email protected] for the room location.

    Abstract:
    Traditional methods for collecting angler behavior data, such as creel and aerial surveys, are costly, while modern alternatives like online platforms and smartphone apps offer cost-effective solutions. Previous studies identified correlations between citizen-reported data and conventional surveys but did not investigate direct relationships between the two or the role of intermediate variables. Using Bayesian networks, we examined these relationships for two key metrics—daily catch rate and fishing pressure—based on data from Alberta and Ontario, Canada. Our analysis included meteorological factors and day types as intermediates, with Bayesian model averaging to assess variable connections. Results indicated moderate direct links between webpage views and aerial boat counts in Ontario and low-probability links between creel survey catch rates and citizen-reported data in Alberta, with indirect links mediated by temperature and solar radiation. We also compared expert-based, ChatGPT-driven, and data-driven Bayesian network models for predicting aerial survey boat counts. All models showed similar predictive accuracy and identified the number of webpage visits as a key predictor of boat activity. These findings highlight the comparable performance of expert, AI-assisted, and data-driven approaches when key variables are considered.