Narges Armanfard

Timeline

1. 2016

   Ph.D. at McMaster University

   She earns a Ph.D. in artificial intelligence at McMaster University in 2016. Her doctoral work develops machine learning algorithms for data classification, among them localized feature selection and a system for automatic and continuous coma patient assessment and outcome prediction. She then does postdoctoral research at the University of Toronto with Alex Mihailidis on machine learning for home-based physiological signal monitoring of elderly adults.

2. 2021

   The DASVDD anomaly detector

   In 2021, she and Hadi Hojjati post DASVDD to arXiv, an anomaly detection method that learns an autoencoder while minimizing the volume of a hypersphere around its latent representation. The paper appears in IEEE Transactions on Knowledge and Data Engineering in 2023.

3. 2023

   C3 contrastive clustering

   In 2023, she coauthors C3 (Cross-instance guided Contrastive Clustering) with Mohammadreza Sadeghi and Hadi Hojjati; the paper is accepted at the 34th British Machine Vision Conference and improves clustering accuracy by 6.6% on CIFAR-10 and 3.3% on CIFAR-100. The iSMART Lab also organizes a tutorial on deep learning for unsupervised time-series anomaly detection at IJCAI 2023.

4. 2024

   Launching the ASTAD workshop

   In 2024, the iSMART Lab organizes the first Workshop on Automated Spatial and Temporal Anomaly Detection (ASTAD) at WACV 2024 and a tutorial on deep learning for unsupervised time-series anomaly detection at AAAI 2024. In March 2024, she describes the lab's work on an AI-based system that identifies the epileptogenic zone from intracerebral EEG to guide epilepsy surgery.

5. 2024

   MOSAD for open-set detection

   In 2024, she coauthors MOSAD with Thomas Lai and Thi Kieu Khanh Ho, accepted at ECAI 2024. MOSAD is a multi-head detector for open-set multivariate time-series anomaly detection that learns from a few labeled anomalies during training.

6. May 2025

   A McMaster distinguished lecture

   In May 2025, she gives the fourth lecture of the ECE Distinguished Research Lecture Series for 2024–2025 at McMaster University, where she earned her Ph.D. Titled "Anomaly Detection in Time Series Data," the talk is a tutorial on deep learning techniques for detecting anomalies in large-scale time series datasets, and she speaks as an Associate Professor at McGill University.

7. 2025–2026

   MultiTab-Net and EngineAD

   MultiTab-Net, a multitask transformer architecture for large tabular data, appears on arXiv in November 2025 with Dimitrios Sinodinos and Jack Yi Wei as coauthors and is accepted at AAAI 2026. The EngineAD dataset, collected from a fleet of 25 commercial vehicles over 6 months, appears at the ASTAD workshop at AAAI 2026.

8. As of October 2026

   Founder of the iSMART Lab

   She is an Associate Professor (tenured) in the Department of Electrical and Computer Engineering at McGill University, an associate academic member at Mila and the founder and principal investigator of the iSMART Lab. Research at the lab covers time-series analysis, computer vision, reinforcement learning and representation learning, with applications in health care and industry. In April 2026, the lab receives an FRQ award for its SeeVita project on contactless vital-signs monitoring, and in June 2026 it posts EpiiSLM, a dual foundation model system for locating the epileptogenic zone from stereo-EEG recordings.