My Research Interests

  • Dependable and Secure Deep Neural Networks
    • Uncertainty Estimation, Detection of adversarial examples & Out-of-Distribution samples
  • Adversarial Networks; Domain Adversarial Networks (DAN), Generative Adversarial Networks (GAN)
  • Computer Vision
    • Robust Object Detection, 3D Human Pose Estimation

  • Self-supervised Robust Object Detectors from Partially Labelled Datasets

    Mahdieh Abbasi, Denis Laurendeau, Christian Gagné; Arxiv, 2020, [PDF][Code](coming soon)

  • Toward Adversarial Robustness by Diversity in an Ensemble of Specialized Deep Neural Networks

    Mahdieh Abbasi, Arezoo Rajabi, Christian Gagne, Rakesh B. Bobba; Long paper in Canadian Conference on AI, 2020( Best paper award). [PDF][Talk] [Code](coming soon)

  • Towards metrics for differentiating Out-of-Distribution sets

    Mahdieh Abbasi, Changjian Shui, Arezoo Rajabi, Christian Gagne, Rakesh B. Bobba; In European Conference on AI, 2020 (Oral; acceptance rate 26.8%). [PDF] [Talk] [Presentation][Code]

  • A Principled Approach for Learning Task Similarity in Multitask Learning

    Changjian Shui, Mahdieh Abbasi, Louis-Emile Robitaille, Boyu Wang, Christian Gagné; In International Joint Conference on Artificial Intelligence , 2019 (Poster, acceptance rate 18%). [PDF] [Poster][Code]

  • Towards Dependable Deep Convolutional Neural Networks (CNNs) with Out-Distribution Learning

    Mahdieh Abbasi, Arezoo Rajabi, Christian Gagné, Rakesh B. Bobba; In Dependable and Secure Machine Learning (DSML), co-located with Dependable System Networks (DSN), 2018. [PDF] [Presentation][Code]

  • Robustness to Adversarial Examples through an Ensemble of Specialists

    Mahdieh Abbasi and Christian Gagné; In International Conference on Learning Representations (ICLR) Workshop, 2017.[PDF] [Poster]

  • Alternating Direction Method of Multipliers for Sparse Convolutional Neural Networks

    Farkhondeh Kiaee, Christian Gagné, Mahdieh Abbasi; In Arxiv, 2017

  • Monocular 3D Human Pose Estimation with a Semi-supervised Graph-based Method

    Mahdieh Abbasi, Hamid Reza Rabiee, Christian Gagné; In 3D Computer Vision IEEE Conference, 2015 (acceptance rate for oral 15%). [PDF] [Poster]