I am currently completing my final year as a PhD candidate in the Department of Computer and Electrical Engineering at the University of Waterloo. My research focuses on optimization methods in machine learning, with particular interests in federated learning and knowledge distillation. Prior to pursuing my Ph.D., I earned my graduate degree from the same department at the University of Waterloo.

My academic journey has equipped me with strong interpersonal skills, and I am enthusiastic about collaborating with others. I have a track record of publications in prestigious AI/ML conferences, including ICLR and TMLR. Additionally, I have gained expertise in programming languages such as Python, and frameworks like PyTorch and Tensorflow.

Recent Peer-Reviewed Papers:

  1. E-H Yang, S.M. Hamidi, L Ye, R Tan, B Yang, "Conditional Mutual Information Constrained Deep Learning: Framework and Preliminary Results", ISIT 2024.
  2. S.M. Hamidi, R Tan, L Ye, E-H Yang, "Fed-IT: Addressing Class Imbalance in Federated Learning through an Information-Theoretic Lens", ISIT 2024.
  3. S.M. Hamidi, "Training Neural Networks on Remote Edge Devices for Unseen Class Classification", IEEE Signal Processing Letters 2024.
  4. S.M. Hamidi*, L Ye*, R Tan*, E-H Yang, "Bayes Conditional Distribution Estimation for Knowledge Distillation Based on Conditional Mutual Information", ICLR 2024.
  5. S.M. Hamidi, E-H Yang, "AdaFed: Fair Federated Learning via Adaptive Common Descent Direction", Transaction Machine Learning Research (TMLR 2024).
  6. S.M. Hamidi, O Damen, "Fair wireless federated learning through the identification of a common descent direction",  IEEE Communications Letters 2024.
  7. S.M. Hamidi, L Ye, "Robustness Against Adversarial Attacks via Learning Confined Adversarial Polytopes", ICASSP 2024.

Patents:

  1. S.M. Hamidi, S Herath, A Bayesteh, AK Khandani, "Systems and methods for communication resource usage control", US Patent App. 15/824,352.
  2. P Struhsaker, P Posner, SM Hamidi, AK Khandani, "Methods for formation of antenna array from sub-arrays", US Patent 10,756,443.