• Class Imbalance in Deep Learning: Impacts and solutions
Published:
This talk presents my latest work on understanding and addressing the problem of class imbalance in deep learning.
Posts by Collection
mirl
news
talks
Reinforcement Learning for Materials Discovery
Published:
This talk focused on the application of reinforcement learning to automated chemistry, materials design and self-driving labs.
Reinforcement Learning for Satellite Communication
Published:
This talk presents my recent work applying reinforcement learning to various applications of satellite communications.
teaching
ISAP3001A Prncples or Apps in Data Analysis
Undergraduate course, Carleton University, Integrated Science, 2021
Data analysis strategies to tackle real-world, wicked problems. Includes a hands-on applied environmental data science project with a variety of partners. Topics include: obtaining and working with data, exploring causal relationships, data ethics, communicating data, and moving from data to information to action.
ISAP3001A Prncples or Apps in Data Analysis
Undergraduate course, Carleton University, Integrated Science, 2022
Data analysis strategies to tackle real-world, wicked problems. Includes a hands-on applied environmental data science project with a variety of partners. Topics include: obtaining and working with data, exploring causal relationships, data ethics, communicating data, and moving from data to information to action.
ISAP3001A Prncples or Apps in Data Analysis
Undergraduate course, Carleton University, Integrated Science, 2023
Data analysis strategies to tackle real-world, wicked problems. Includes a hands-on applied environmental data science project with a variety of partners. Topics include: obtaining and working with data, exploring causal relationships, data ethics, communicating data, and moving from data to information to action.
ITI 1121 Introduction to Computing II
Undergraduate course, University of Ottawa, Information Technology, 2026
Object-oriented programming. Abstraction principles: information hiding and encapsulation. Linked lists, stacks, queues, binary search trees. Iterative and recursive processing of data structures. Virtual machines.
CSI 5340 Introduction to Deep Learning and Reinforcement Learning
Graduate course, University of Ottawa, Computer Science, 2026
Fundamental of machine learning; multi-layer perceptron, universal approximation theorem, back-propagation; convolutional networks, recurrent neural networks, variational auto-encoder, generative adversarial networks; components and techniques in deep learning; Markov Decision Process; Bellman equation, policy iteration, value iteration, Monte-Carlo learning, temporal difference methods, Q-learning, SARSA, applications. This course is equivalent to COMP 5340 at Carleton University.