Ahmad Sajedi

I am Ahmad Sajedi, currently a Machine Learning Engineer II at Instacart. Before this role, I completed my Ph.D. at the University of Toronto, where I was fortunate to be advised by Prof. Konstantinos N. Platanioits and Prof. Yuri A. Lawryshyn. Prior to joining UofT, I received my M.Sc. degree from the Electrical and Computer Engineering Department at the University of Waterloo, mentored by Prof. En-Hui Yang.

My research primarily focuses on Efficient Learning, Fraud Prevention, and Visual Multilabel Representations. I am open to collaboration and available to address any inquiries regarding my research. Please do not hesitate to reach out to me via email.

Email: sajedi [dot] ah [at] gmail [dot] com

Email  |  Resume  |  LinkedIn  |  GitHub

profile photo

News

• [Oct. 2024]: Started at Instacart 🥕 as a Machine Learning Engineer II.
• [Sep. 2024]: Successfully passed my final Ph.D. oral examination and officially awarded my Ph.D. degree.
• [Jul. 2024]: Data-to-Model Distillation paper has been accepted by ECCV 2024. Paper, code, and webpage coming soon.
• [May. 2024]: Passed my Ph.D. Departmental Oral Examination. My thesis is nominated for an award!
• [May. 2024]: My invention disclosures for two patents, DataDAM and D2M, have been accepted as complete.
• [Apr. 2024]: ATOM has been accepted by CVPR-DD 2024.
• [Apr. 2024]: The First Dataset Distillation Challenge has been accepted by ECCV 2024. Serving as a Primary Chair.
• [Apr. 2024]: Give a talk at Royal Bank of Canada (RBC), invited by Prof. Yuri A. Lawryshyn. Slides coming soon.
• [Dec. 2023]: ProbMCL has been accepted by ICASSP 2024.
• [Jul. 2023]: DataDAM has been accepted by ICCV 2023.
• [May. 2023]: Give a talk at Royal Bank of Canada (RBC), invited by Prof. Yuri A. Lawryshyn. Slides coming soon.
• [Feb. 2023]: A New Probabilistic Distance has been accepted by ICASSP 2023.
• [May. 2022]: SKD has been accepted by IVMSP 2022.
• [Dec. 2021]: Passed my Ph.D. Thesis Proposal defense!

Publications & Manuscripts

b3do Data-to-Model Distillation: Data-Efficient Learning Framework
Ahmad Sajedi, Samir Khaki, Lucy Z. Liu, Ehsan Amjadian, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
ECCV, 2024
Website (Coming Soon) | Paper | Code (Coming Soon)

We present D2M, which embeds knowledge into a generative model, allowing for efficient and scalable training across various distillation ratios and architectures.

b3do ATOM: Attention Mixer for Efficient Dataset Distillation
Samir Khaki, Ahmad Sajedi, Kai Wang, Lucy Z. Liu, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
CVPR-DD, 2024
Website (Coming Soon) | Paper | Code (Coming Soon)

We introduce the ATOM module to leverage contextual and localization information from the channel and spatial attention mechanisms to improve dataset distillation efficiency.

b3do ProbMCL: Simple Probabilistic Contrastive Learning For Multi-label Visual Classification
Ahmad Sajedi, Samir Khaki, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
ICASSP, 2024
Website | Paper

We propose a simple yet effective probabilistic contrastive learning framework for multi-label image classification tasks using Gaussian mixture models.

b3do DataDAM: Efficient Dataset Distillation with Attention Matching
Ahmad Sajedi, Samir Khaki, Ehsan Amjadian, Lucy Z. Liu, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
ICCV, 2023
Website | Paper | Code

We present an effective learning framework to distill informative knowledge from a large-scale training dataset into a small, synthetic one.

b3do A New Probabilistic Distance Metric With Application In Gaussian Mixture Reduction
Ahmad Sajedi, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
ICASSP, 2023
Paper

We propose a new probabilistic distance metric to compare two continuous probability density functions. This metric provides a closed-form expression for Gaussian Mixture Models.

b3do End-to-End Supervised Multilabel Contrastive Learning
Ahmad Sajedi, Samir Khaki, Konstantinos N. Plataniotis, Mahdi S. Hosseini
arXiv, 2023
Paper | Code

We present an end-to-end kernel-based contrastive learning framework designed for multilabel datasets in computer vision and medical imaging.

b3do FedPnP: Personalized Graph-Structured Federated Learning
Arash Rasti-Meymandi, Ahmad Sajedi, Konstantinos N. Plataniotis
ECCV preprint, 2024
Paper

We introduce a novel personalized federated learning algorithm that leverages the inherent graph-based relationships among clients.

b3do Subclass Knowledge Distillation with Known Subclass Labels
Ahmad Sajedi, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
IVMSP, 2022
Paper

We present subclass knowledge distillation, a method of transferring predicted subclass knowledge from a teacher to a smaller student model, designed for clinical applications.

b3do On the Efficiency of Subclass Knowledge Distillation in Classification Tasks
Ahmad Sajedi, Konstantinos N. Plataniotis
arXiv, 2022
Paper

We introduce a novel model distillation algorithm that leverages subclass labels' knowledge and quantifies the information the teacher can provide to the student through our framework.

b3do High-Performance Convolution Using Sparsity and Patterns for Inference in Deep Convolutional Neural Networks
Hossam Amer, Ahmed H. Salamah, Ahmad Sajedi, En-hui Yang
arXiv, 2021
Paper | Code

By leveraging feature map sparsity, we introduce two novel convolution algorithms aiming to reduce memory usage, enhance inference speed, and maintain accuracy simultaneously.

Patents

b3do Efficient Dataset Distillation with Attention Matching
US Patent

Ahmad Sajedi, Ehsan Amjadian, Samir Khaki, Lucy Z. Liu, Yuri A. Lawryshyn, Konstantinos N. Plataniotis

b3do Data-to-Model Distillation
US Patent

Ahmad Sajedi, Ehsan Amjadian, Samir Khaki, Lucy Z. Liu, Yuri A. Lawryshyn, Konstantinos N. Plataniotis

b3do Tabular Dataset Condensation
US Patent Pending

Samir Khaki, Ahmad Sajedi, Lucy Z. Liu, Yuri A. Lawryshyn, Konstantinos N. Plataniotis

Education

University of Toronto, Toronto, Canada
Ph.D. in Electrical and Computer Engineering • Sept. 2020 to Sept. 2024
University of Waterloo, Waterloo, Canada
M.Sc. in Electrical and Computer Engineering • Sept. 2018 to Aug. 2020
Amirkabir University of Technology, Tehran, Iran
B.Sc. in Electrical and Computer Engineering • Sept. 2014 to Aug. 2018

Experiences

Instacart
Machine Learning Engineer II • Oct. 2024 to Present
Manager: Dr. Sophia Li
Royal Bank of Canada (RBC)
Machine Learning Researcher • Jan. 2022 to Sep.2024
Mentor: Dr. Lucy Z. Liu and Dr. Ehsan Amjadian
Centre for Management of Technology & Entrepreneurship (CMTE), University of Toronto
Graduate Research Associate • Sept. 2021 to Sept. 2024
Advisor: Prof. Yuri A. Lawryshyn
Multimedia Lab (Bell), University of Toronto
Graduate Research Associate • Sept. 2020 to Sept. 2024
Advisor: Prof. Konstantinos N. Plataniotis
Multimedia Communications Lab (Leitch), University of Waterloo
Graduate Research Assistant • Sept. 2018 to Aug. 2020
Advisor: Prof. En-Hui Yang

Professional Academic Activities

Primary Chair: • The First Dataset Distillation Challenge Workshop at ECCV 2024
Reviewer: • ICLR 2025 • NeurIPS 2024 • ECCV 2024 • ICASSP 2024

Selected Teaching

Teaching Assistant • ECE1512, Digital Image Processing and Applications • Fall 2023/2022, Winter 2022
Teaching Assistant • MIE1626, Data Science Methods and Statistical Learning • Fall 2023/2022, Winter 2024/2022
Teaching Assistant • ECE602, Convex Optimization • Summer 2021, Winter 2020
Teaching Assistant • ECE302, Probability and Applications • Fall 2023/2022/2021/2020, Winter 2024/2022
Teaching Assistant • ECE286, Probability and Statistics • Winter 2024/2023/2022/2021
Teaching Assistant • STA237, Probability, Statistics, and Data Analysis • Fall 2023/2021

Source code from Jon Barron's lovely website.