Dr. Canh T. Dinh is a Machine Learning Engineer at Canva (Australia) and holds a PhD from the University of Sydney. His research focuses on federated learning, optimization, machine learning, and hardware acceleration. He is best known for his work on Personalized Federated Learning with Moreau Envelopes, a highly cited NeurIPS 2020 paper that has influenced both the theoretical and practical development of personalized distributed learning systems.
His contributions span wireless network-aware federated learning, multi-task federated optimization with Laplacian regularization, and federated edge learning acceleration. Dr. Dinh’s work has been published in premier venues such as NeurIPS, IEEE/ACM Transactions on Networking, and IEEE Transactions on Neural Networks and Learning Systems, with applications ranging from large-scale AI model personalization to FPGA-based system design. His research has been cited over 2000 times, reflecting his significant impact on the field.