Selected Publications

Responsible machine learning (Fairness)

  • Intersectional Unfairness Discovery.
    Gezheng Xu, Qi Chen, Charles Ling, Boyu Wang, Changjian Shui.
    International Conference on Machine Learning (ICML), 2024.
    [Paper] [Code]

  • On Learning Fairness and Accuracy on Multiple Subgroups.
    Changjian Shui, Gezheng Xu (Equal Contribution), Qi Chen, Jiaqi Li, Charles Ling, Tal Arbel, Boyu Wang, Christian Gagné.
    Neural Information Processing Systems (NeurIPS) 2022.
    [Openreview] [Code] [Video] [Spotlight Video]

  • Fair Representation Learning through Implicit Path Alignment.
    Changjian Shui, Qi Chen, Jiaqi Li, Boyu Wang, Christian Gagné.
    International Conference on Machine Learning (ICML) 2022.
    [Paper] [Code] [Video] [Slides]

  • Mitigating Calibration Bias Without Fixed Attribute Grouping for Improved Fairness in Medical Imaging Analysis. [Paper]
    Changjian Shui, Justin Szeto, Raghav Mehta, Douglas Arnold, Tal Arbel.
    Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2023 (Early acceptance, 14% of submissions)

Reliable machine learning via distribution shift (Robustness, Out-of-distribution generalization)

  • On the Benefits of Representation Regularization in Invariance based Domain Generalization.
    Changjian Shui, Boyu Wang, and Christian Gagné.
    Machine Learning Journal (MLJ) 2022.
    [Paper] [Video] [Simple Code Demo]

  • On the Stability-Plasticity Dilemma in Continual Meta-Learning: Theory and Algorithm.
    Qi Chen, Shui Changjian, Ligong Han, Mario Marchand.
    [Openreview]
    Neural Information Processing Systems (NeurIPS) 2023

  • Generalization Bounds For Meta-Learning: An Information-Theoretic Analysis.
    Qi Chen, Changjian Shui, Mario Marchand
    Neural Information Processing Systems (NeurIPS) 2021. (Spotlight, 3% of submissions)
    [Openreview]

  • Aggregating From Multiple Target-Shifted Sources.
    Changjian Shui, Zijian Li, Jiaqi Li, Christian Gagné, Charles Ling, Boyu Wang.
    International Conference on Machine Learning (ICML) 2021.
    [Paper] [Code] [Slides] [Video]

  • A principled approach for learning task similarity in multitask learning.
    Changjian Shui, Mahdieh Abbasi, Louis-Émile Robitaille, Boyu Wang, Christian Gagné
    International Joint Conference on Artificial Intelligence (IJCAI) 2019.
    [Paper] [Code] [Slides]

  • Deep Active Learning: Unified and Principled Method for Query and Training.
    Changjian Shui, Fan Zhou, Christian Gagné, Boyu Wang
    International Conference on Artificial Intelligence and Statistics (AISTATS) 2020.
    [Paper] [Code] [Slides] [Video]

Recent Publications

2024

  • Learning Equilibrium Transformation for Gamut Expansion and Color Restoration.
    Jun Xiao, Changjian Shui, Zhi-Song Liu, Qian Ye, and Kin-Man Lam
    European Conference on Computer Vision (ECCV), 2024

  • Towards Progressive Multi-Frequency Representation for Image Warping.
    Jun Xiao, Zihang Lyu, Cong Zhang, Yakun Ju, Changjian Shui, and Kin-Man Lam.
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024

  • Generalizing Across Temporal Domains with Koopman Operators.
    Qiuhao Zeng, Wei Wang, Fan Zhou, Gezheng Xu, Ruizhi Pu, Changjian Shui, Christian Gagné, Shichun Yang, Charles Ling, and Boyu Wang.
    AAAI Conference on Artificial Intelligence (AAAI), 2024

2023

  • On the Stability-Plasticity Dilemma in Continual Meta-Learning: Theory and Algorithm.
    Qi Chen, Shui Changjian, Ligong Han, Mario Marchand.
    [Openreview]
    Neural Information Processing Systems (NeurIPS) 2023

  • Mitigating Calibration Bias Without Fixed Attribute Grouping for Improved Fairness in Medical Imaging Analysis. [Paper]
    Changjian Shui, Justin Szeto, Raghav Mehta, Douglas Arnold, Tal Arbel.
    Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2023 (Early acceptance, 14% of submissions)

  • Label Shift Conditioned Hybrid Querying for Deep Active Learning, [Paper]
    Jiaqi Li, Haojia Kong, Gezheng Xu, Changjian Shui, Ruizhi Pu, Zhao Kang, Charles X. Ling, and Boyu Wang.
    Knowledge-Based Systems (KBS) 2023.

  • Towards More General Loss and Setting in Unsupervised Domain Adaptation. [Paper]
    Changjian Shui, Ruizhi Pu (Equal Contribution), Gezheng Xu, Jun Wen, Fan Zhou, Christian Gagné, Charles X. Ling, and Boyu Wang.
    IEEE Transactions on Knowledge and Data Engineering (TKDE) 2023.

  • Evaluating the Fairness of Deep Learning Uncertainty Estimates in Medical Image Analysis. [Paper]
    Mehta, Raghav, Changjian Shui, and Tal Arbel.
    Conference on Medical Imaging with Deep Learning (MIDL) 2023.

  • Gap Minimization for Knowledge Sharing and Transfer. [arxiv]
    Boyu Wang, Jorge Mendez, Changjian Shui, Fan Zhou, Di Wu, Gezheng Xu, Christian Gagné, and Eric Eaton.
    Journal of Machine Learning Research (JMLR) 2023.

2022

  • On Learning Fairness and Accuracy on Multiple Subgroups.
    Changjian Shui, Gezheng Xu (Equal Contribution), Qi Chen, Jiaqi Li, Charles Ling, Tal Arbel, Boyu Wang, Christian Gagné.
    Neural Information Processing Systems (NeurIPS) 2022.
    [Openreview] [Code] [Video] [Spotlight Video]

  • Fair Representation Learning through Implicit Path Alignment.
    Changjian Shui, Qi Chen, Jiaqi Li, Boyu Wang, Christian Gagné.
    International Conference on Machine Learning (ICML) 2022.
    [Paper] [Code] [Video] [Slides]

  • On the Benefits of Representation Regularization in Invariance based Domain Generalization.
    Changjian Shui, Boyu Wang, and Christian Gagné.
    Machine Learning Journal (MLJ) 2022.
    [Paper] [Video] [Simple Code Demo]

  • Clinically Plausible Pathology-Anatomy Disentanglement in Patient Brain MRI with Structured Variational Priors [arXiv]
    Hu, Anjun, Falet, Jean-Pierre R, Nichyporuk, Brennan S, Shui, Changjian, Arnold, Douglas L, Tsaftaris, Sotirios A, Arbel, Tal.
    ML4H short paper

  • Lifelong Online Learning from Accumulated Knowledge.
    Changjian Shui, Wei Wang, Ihsen Hedhli, Feng Wang, Boyu Wang, and Christian Gagné.
    ACM Transactions on Knowledge Discovery from Data (TKDD) 2022.

  • Information Gain Sampling for Active Learning in Medical Image Classification
    Raghav Mehta, Changjian Shui, Brennan Nichyporuk, and Tal Arbel
    Uncertainty for Safe Utilization of Machine Learning in Medical Imaging (UNSURE) workshop. MICCAI 2022. [Paper]

  • Evolving Domain Generalization. [arxiv]
    William Wei Wang, Gezheng Xu, Ruizhi Pu, Jiaqi Li, Fan Zhou, Changjian Shui, Charles Ling, Christian Gagné, Boyu Wang.
    Arxiv 2022.

  • A Novel Domain Adaptation Theory With Jensen-Shannon Divergence.
    Changjian Shui, Qi Chen, Jun Wen, Fan Zhou, Christian Gagné, and Boyu Wang.
    Knowledge-Based Systems (KBS) 2022. [Preliminary version]

Prior to 2022

  • Generalization Bounds For Meta-Learning: An Information-Theoretic Analysis.
    Qi Chen, Changjian Shui, Mario Marchand
    Neural Information Processing Systems (NeurIPS) 2021. (Spotlight, 3% of submissions)
    [Openreview]

  • Aggregating From Multiple Target-Shifted Sources.
    Changjian Shui, Zijian Li, Jiaqi Li, Christian Gagné, Charles Ling, Boyu Wang.
    International Conference on Machine Learning (ICML) 2021.
    [Paper] [Code] [Slides] [Video]

  • On the Benefits of Two Dimensional Metric Learning.
    Di Wu, Fan Zhou, Boyu Wang, Chi Man Wong, Changjian Shui, Yuan Zhou, Qicheng Lao, Feng Wan
    IEEE Transactions on Knowledge and Data Engineering (TKDE) 2021. [Paper]

  • Common Spatial Pattern Reformulated for Regularizations in Brain-Computer Interfaces.
    Boyu Wang, Chi Man Wong, Zhao Kang, Feng Liu, Changjian Shui, Feng Wan, and C. L. Philip Chen
    IEEE Transactions on Cybernetics (TCYB) 2021. [paper]

  • Task similarity estimation through adversarial multitask neural network.
    Fan Zhou, Changjian Shui, Mahdieh Abbasi, Louis-Émile Robitaille, Boyu Wang, Christian Gagné.
    IEEE Transcations on Neural Networks and Learning Systems (TNNLS) 2021. [Paper]

  • Domain Generalization via Optimal Transport with Metric Similarity Learning.
    Fan Zhou, Zhuqing Jiang, Changjian Shui, Boyu Wang, Brahim Chaib-draa.
    Neurocomputing 2021. [Paper]

  • Discriminative Active Learning for Domain Adaptation.
    Fan Zhou, Changjian Shui, Shichun Yang, Bincheng Huang, Boyu Wang, and Brahim Chaib-draa.
    Knowledge-Based Systems (KBS) 2021. [Paper]

  • Deep Active Learning: Unified and Principled Method for Query and Training.
    Changjian Shui, Fan Zhou, Christian Gagné, Boyu Wang
    International Conference on Artificial Intelligence and Statistics (AISTATS) 2020.
    [Paper] [Code] [Slides] [Video]

  • Toward Metrics for Differentiating Out-of-Distribution Sets
    Mahdieh Abbasi, Changjian Shui, Arezoo Rajabi, Christian Gagne, Rakesh Bobba.
    European Conference on Artificial Intelligence (ECAI) 2020. [Paper]

  • A principled approach for learning task similarity in multitask learning.
    Changjian Shui, Mahdieh Abbasi, Louis-Émile Robitaille, Boyu Wang, Christian Gagné
    International Joint Conference on Artificial Intelligence (IJCAI) 2019.
    [Paper] [Code] [Slides]

  • Diversity regularization in deep ensembles. [arxiv]
    Changjian Shui, Azadeh Sadat Mozafari, Jonathan Marek, Ihsen Hedhli, Christian Gagné.
    Arxiv