Majid Komeili

Research

Research Interests

My research interest is in machine learning, deep learning and related areas in computer vision and natural language processing. More specifically, I am interested in interpretable machine learning, deep neural networks and transfer learning. The goal is to create ML models with explainability in mind, or to develop methods that can decipher existing black-box ML models. I am also interested in how ML models can learn to perform new tasks with limitted amount of labeled data; a capability that human is very good at. Along this line, generative models and effective use of language models and vision-language models are crucial.

Current projects

• Explainable AI and NLP to help better understand issues around misinformation and vaccine hesitancy in social media (collaboration with the Department of law and Legal Studies and the School of Journalism and Communication at Carleton U)
• Explainable AI and NLP for assessment of functional limitations and disability services for postsecondary education (collaboration with the Accessibility Institute)
• Generative models for communicating graphical information to visually impaired or blind individuals
• Explainable AI for predicting chronic homelessness (collaboration with the City of Ottawa)
• Digital tools for revitalizing endangered languages (ELK-Tech)

Past projects

• Explainable AI for predictive analytics in employee benefits insurance
• Biometrics, spoofing attacks and countermeasures
• ML for analyzing brain signals (EEG)
• ML for unobtrusive monitoring of vital physiologic parameters
• ML for analyzing brain MRI images in patients with ASD

Students

Current members

Abbas Akkasi, Postdoctoral Fellow (with Boris Vukovic and Kathleen Fraser)

Mohammad Reza Zarei, PhD (with Frank Dehne)

Adnan Khan, PhD

Rakshil Kevadiya, MCS (with Boris Vukovic and Kathleen Fraser)

Alireza Choubineh, MCS

Hoda Vafaeesefat, MCS

Youssef Fahmani, MCS (with Adrian Chan)

Past Grad Students

Mitchell Chatterjee, MCS (with Adrian Chan)

Seyed Omid Davoudi, PhD, Winter 2024, (with Frank Dehne), moved on to Larus Technologies

Aatreyi Pranavbhai Mehta, MCS, Winter 2023, moved on to Razor Sharp Consulting

Galen O'Shea, MCS, Winter 2023, moved on to Mission Control

Mohammad Mahdi Heydari Dastjerdi, MCS, Summer 2022, Paphus Solutions

Mohammad Nokhbeh Zaeem, MCS, Winter 2021, moved on to SoundHound Inc

Siraj Ahmed, MCS, Fall 2020, U Ottawa, Co-supervised with Prof. J. Park, moved on to Braiyt AI Inc

Abhijeet Chauhan, MCS, Summer 2020, moved on to IMRSV Data Labs

Past Undergrad Students

Saurabh Gummaraj Kishore, Honors Project

David Hobson, Winter 2023, Honors Thesis

Kailash Balakrishnan, Winter 2023, Honors Project

Jesse Mendoza, Fall 2022, Honors Project

Hilaire Djani, Winter 2022, Honors Thesis

Tim Elliott, Honors Project

Juntong He, Honors Project

Qixiang Luan, Honors Project

M. Kazman, Fall 2021, Honors Project.

A. Ong, Fall 2021, Honors Project.

J. Woo, Summer 2021, Honors Project.

I. Nicolaev, Summer 2021, Honors Project.

M. Kazman, Summer 2021, Honors Project.

J. Geng, Winter 2021, Honors Project.

Y. Song, Winter 2021, Honors Project.

K. Zhen, Winter 2021, Honors Project.

H. Le, Fall 2020, Honors Project.

Y. Gao, Fall 2020, Honors Project.

T. Cao, Fall 2020, Honors Project.

Y. Chen, Fall 2020, Honors Project.

V. Nguyen, Summer 2020, Honors Project.

J. Danovitch, Winter 2020, Honors Thesis.

M. Kuzmenko, Winter 2020, Honors Project.

L. Wise, Winter 2020, Honors Project.

L. Koftinow-Mikan, Fall 2019, Honors Project.

X. Liu, Fall 2019, Honors Project.

G. O'Shea, Summer 2019, Honors Project.

L. Colwell, Summer 2019, DSRI internship.

K. Causton, Summer 2019, Honors Project (with Oliver).

Y. Yamanaka, Winter 2019, Honors Project.

S. Kudolo, Winter 2019, Honors Project.

L. Gruska, Winter 2019, Honors Project.

L. He, Winter 2019, Honors Project.

Joining/Volunteering

Applying for MSc or PhD:

MSc and PhD applicants who are interested in my research are encouraged to contact me via email.

Prerequisites: A good candidate should have background in probability and linear algebra, and have had courses in Machine Learning or related areas including Computer Vision and Natural Language Processing.

Prospective MSc and PhD students who are applying to the School of Computer Science at Carleton University and are interested in my research are encouraged to indicate my name as their preferred research supervisor.

Please note that due to the volume of emails I receive, I am not able to respond to all.

Undergrad students at Carleton University who are interested in doing their Honours project/thesis with me, are encouraged to contact me via email.

Teaching

COMP 5801: Advanced Machine Learning - Fall 2024.

COMP 4102: Computer Vision - Winter 2024.

DATA 5000: Introduction to Data Science - Winter 2024.

COMP 5900: Advanced Machine Learning - Fall 2023.

DATA 5000: Introduction to Data Science - Winter 2023.

COMP 5900: Advanced Machine Learning - Fall 2022.

DATA 5000: Introduction to Data Science - Winter 2022.

COMP 5900: Advanced Machine Learning - Fall 2021.

COMP 4105: Introduction to Machine Learning - Fall 2021.

COMP 5900V: Advanced Machine Learning - Winter 2021.

DATA 5000: Introduction to Data Science - Winter 2021.

COMP 4900A: Introduction to Machine Learning - Fall 2020.

COMP 4900B: Introduction to Machine Learning - Winter 2020.

DATA 5000: Introduction to Data Science - Winter 2020.

COMP 5900X: Advanced Machine Learning - Fall 2019.

COMP 5900: Machine Learning for Healthcare - Winter 2019.

Publications

Selected Publications

M. R. Zarei, M. Komeili (2024), “Interpretable Few-shot Learning with Online Attribute Selection”, Neurocomputing pdf

O. Davoodi, M. Komeili (2024), “Feature-based explainable reinforcement learning in environments with multiple sources of risk”, International Conference on Pattern Recognition and Artificial Intelligence

A. Akkasi, A. Khan, M. Shaaban, M. Komeili, and M. Yaqub, (2024) “iML at Semeval-2024 task 2: Safe biomedical natural language inference for clinical trials with LLM based ensemble inferencing”, 18th ACL International Workshop on Semantic Evaluation.

M. A. Choukali, M. Chehel-Amirani, M. Valizadeh, A. Abbasi and M. Komeili (2024), “Pseudo-Class Part Prototype Networks for Interpretable Breast Cancer Classification”, Nature Scientific Reports. pdf

A. Akkasi, K. C. Fraser, and M. Komeili (2023), “Reference-free Summarization Evaluation with Large Language Models”, IJCNLP-AACL Workshop on Evaluation and Comparison for NLP systems.

M. Shaaban, A. Akkasi, A. Khan, M. Komeili, M. Yaqub (2023), “Fine-Tuned Large Language Models for Symptom Recognition from Spanish Clinical Text”, BioCreative.pdf

O. Davoodi, S. Mohammadizadehsamakosh, M. Komeili (2023), “On the Interpretability of Part-Prototype Based Classifiers: A Human Centric Analysis”, Nature Scientific Reports. pdf

G. O'Shea, M. Komeili (2023), “SuperVision: Self-Supervised Super-Resolution for Appearance-Based Gaze Estimation”, NeurIPS 2023 Gaze Meets ML Workshop. pdf

M. R. Zarei, M. Christensen, S. Everts, M. Komeili (2023), “Vax-Culture: A Dataset for Studying Vaccine Discourse on Twitter”, International Joint Conference on Neural Networks. Preprint

M.R. Zarei, M. Komeili, (2022), “Interpretable Concept-based Prototypical Networks for Few-Shot Learning”, ICIP. pdf

O. Davoudi, M. Komeili, (2022), “Toward Faithful Case-based Reasoning through Learning Prototypes in a Nearest Neighbor-friendly Space”, ICLR. pdf

S. Ahmed, M. Komeili, J. Park, (2022), “Predictive Modelling of Parkinson’s Disease Progression Based on RNA-Sequence with Densely Connected Deep Recurrent Neural Networks”, Nature Scientific Reports. pdf

M. Nokhbeh Zaeem, M. Komeili, (2022), “Cause and Effect: Hierarchical Concept-based Explanation of Neural Networks”, arXiv:2105.07033. preprint

O. Davoudi, M. Komeili, (2021), “Feature-Based Interpretable Reinforcement Learning based on State-Transition Models”, IEEE International Conference on Systems, Man, and Cybernetics, pdf

K. C. Fraser, M. Komeili, (2021), “Measuring Cognitive Status from Speech in a Smart Home Environment”, IEEE Instrumentation and Measurement Magazine, pdf.

M. Nokhbeh Zaeem, M. Komeili, (2021), “Cause and Effect: Concept-based Explanation of Neural Networks”, IEEE International Conference on Systems, Man, and Cybernetics, pdf, code

A. Chauhan, O. Davoudi M. Komeili, (2021), “Multi-scale Deep Nearest Neighbors”, International Joint Conference on Neural Networks. pdf

M. Komeili, N. Armanfard, D. Hatzinakos, (2020), “Multiview Feature Selection for Single-view Classification”, IEEE Transactions on Pattern Analysis and Machine Intelligence. pdf, code

M. Komeili, C. Pou-Prom, D. Liaqat, K. C. Fraser, M. Yancheva, F. Rudzicz, (2019), “Talk2Me: Automated Linguistic Data Collection for Personal Assessment”, PLoS One, pdf, code.

N Armanfard, M. Komeili, JP Reilly, J Connoly, (2019), “A Machine Learning Framework for Automatic and Continuous MMN Detection with Preliminary Results for Coma Outcome Prediction”, IEEE Journal of Biomedical and Health Informatics, vol. 23, issue 4, pp. 1794 -1804.

S. J. Haghighi, M. Komeili, D. Hatzinakos, H. El-Beheiry, (2018), “40-Hz ASSR for Measuring Depth of Anaesthesia During Induction Phase”, IEEE Journal of Biomedical and Health Informatics, vol. 22, issue 6, pp. 1871 - 1882.

M. Komeili, N. Armanfard, D. Hatzinakos, (2018), “Liveness Detection and Automatic Template Updating using Fusion of ECG and Fingerprint”, IEEE Transactions on Information Forensics & Security, vol. 13, issue 7, pp. 1810 - 1822. pdf.

N. Armanfard, J. P. Reilly, M. Komeili, (2018), “Logistic Localized Modeling of the Sample Space for Feature Selection and Classification”, IEEE Transactions on Neural Networks and Learning Systems, vol. 29, issue 5, pp. 1396 - 1413.

N. Armanfard, M. Komeili, A. Mihailidis (2018), “Development of a Smart Home Package for Unobtrusive physiological Monitoring”, IEEE. 40th International Engineering in Medicine and Biology Conference, United States.

A. Kushki, M. Komeili, S. Panahandeh, E. Anagnostou, J. Lerch. (2018), “Examining Associations Between Brain Morphology and Social Function in ASD, ADHD, OCD, and typical development using Machine Learning: Analysis of POND Network Data”, INSAR 2018, International Society for Autism Research, Netherlands.

M. Komeili, W. Louis, N. Armanfard, D. Hatzinakos, (2017), “Feature Selection for Non-stationary Data: Application to Human Recognition using Medical Biometrics”, IEEE Transactions on Cybernetics, vol. 48, issue 5, pp. 1446 - 1459. pdf

N. Armanfard, J. P. Reilly, M. Komeili, (2016), “Local Feature Selection for Data Classification”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 6, pp. 1217-1227.

W. Louis, M. Komeili, D. Hatzinakos, (2016), “Continuous Authentication using One Dimensional Multi-Resolution Local Binary Patterns”, IEEE Transactions on Information Forensics & Security, vol. 11, no. 12, pp 2818-2832. pdf

S. J. Haghighi, M. Komeili, D. Hatzinakos, (2016), “Predicting the Depth of Anaesthesia with 40-Hz ASSR”, IEEE 29th Canadian Conference on Electrical and Computer Engineering (CCECE), Vancouver, Canada.

N. Armanfard, M. Komeili, J. P. Reilly, John F. Connolly, (2016), “Automatic and continuous assessment of ERPs for Mismatch Negativity detection”, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), , Orlando, FL, USA.

M. Komeili, W. Louis, N. Armanfard, D. Hatzinakos, (2016), “Human Recognition using Electrocardiogram Signals: From Rest to Exercise”, IEEE 29th Canadian Conference on Electrical and Computer Engineering (CCECE), Vancouver, Canada.

N. Armanfard, M. Komeili, J. P. Reilly, L. Pino, (2016), “Vigilance lapse identification using sparse EEG electrode arrays”, IEEE 29th Canadian Conference on Electrical and Computer Engineering (CCECE), Vancouver, Canada.

W. Louis, M. Komeili, D. Hatzinakos, (2016), “Real-time Heartbeat Outlier Removal in Electrocardiogram (ECG) Biometric System”, Electrical and Computer Engineering (CCECE), IEEE 29th Canadian Conference on, Vancouver, Canada.

M. Komeili, N. Armanfard, D. Hatzinakos, A. N. Venetsanopoulos, (2015), “Feature Selection from Multisession Electrocardiogram Signals for Identity Verification”, Electrical and Computer Engineering (CCECE), 2015 IEEE 28th Canadian Conference on, Halifax, Canada.

N. Armanfard, M. Komeili, E. Kabir, (2012), “TED: A Texture-Edge Descriptor for Pedestrian Detection in Video Sequences”, Pattern Recognition, vol. 45, no.3, pp. 983-992.

Patents

M. Komeili, N. Armanfard, D. Hatzinakos, “An Expert System for Fingerprint Spoof Detection” International application number CA2019050141, Patent Cooperation Treaty (PCT), Feb. 2019.

N. Armanfard, M. Komeili, J. P. Reilly, John F. Connolly, “Expert System for Automatic, Continuous Coma Patient Assessment and Outcome Prediction” U.S. Provisional Patent, USPTO serial no. 62/509,986, May 2017.