Bo Chang



I am a software engineer at Google Brain, based in Toronto, Canada. Prior to that, I was a machine learning researcher at Borealis AI. I finished my Ph.D. in statistics at the University of British Columbia.

Email:

Education

Employment

Recommender Systems

  • Recency Dropout for Recurrent Recommender Systems
    Bo Chang, Can Xu, Matthieu Lê, Jingchen Feng, Ya Le, Sriraj Badam, Ed H. Chi, Minmin Chen
    Preprint, 2022.
    [arXiv] [BibTeX]

    @article{chang2022recency,
      title={Recency Dropout for Recurrent Recommender Systems},
      author={Chang, Bo and Xu, Can and L{\^e}, Matthieu and Feng, Jingchen and Le, Ya and Badam, Sriraj and Chi, Ed and Chen, Minmin},
      journal={arXiv preprint arXiv:2201.11016},
      year={2022}
    }

  • Learning to Augment for Casual User Recommendation
    Jianling Wang, Ya Le, Bo Chang, Yuyan Wang, Ed H. Chi, Minmin Chen
    ACM Web Conference (WWW), 2022.
    [Proceedings] [BibTeX]

    @inproceedings{wang2022learning,
      title={Learning to Augment for Casual User Recommendation},
      author={Wang, Jianling and Le, Ya and Chang, Bo and Wang, Yuyan and Chi, Ed H and Chen, Minmin},
      booktitle={Proceedings of the ACM Web Conference 2022},
      pages={2183--2194},
      year={2022}
    }

  • User Response Models to Improve a REINFORCE Recommender System
    Minmin Chen, Bo Chang, Can Xu, Ed H. Chi
    ACM International Conference on Web Search and Data Mining (WSDM), 2021.
    [Proceedings] [BibTeX]

    @inproceedings{chen2021user,
      title={User Response Models to Improve a REINFORCE Recommender System},
      author={Chen, Minmin and Chang, Bo and Xu, Can and Chi, Ed H},
      booktitle={Proceedings of the 14th ACM International Conference on Web Search and Data Mining},
      pages={121--129},
      year={2021}
    }

Machine Learning

  • Convolutional Neural Networks Combined with Runge–Kutta Methods
    Mai Zhu, Bo Chang, Chong Fu
    Neural Computing and Applications, 2022.
    [arXiv] [Journal] [BibTeX]

    @article{zhu2022convolutional,
      title={Convolutional neural networks combined with {R}unge--{K}utta methods},
      author={Zhu, Mai and Chang, Bo and Fu, Chong},
      journal={Neural Computing and Applications},
      year={2022},
      publisher={Springer}
    }

  • CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks
    Jiaqi Ma, Bo Chang, Xuefei Zhang, Qiaozhu Mei
    International Conference on Learning Representations (ICLR), 2021.
    [arXiv] [OpenReview] [BibTeX]

    @inproceedings{ma2021copulagnn,
      title={Copula{GNN}: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks},
      author={Ma, Jiaqi and Chang, Bo and Zhang, Xuefei and Mei, Qiaozhu},
      booktitle={International Conference on Learning Representations},
      year={2021},
      url={https://openreview.net/forum?id=XI-OJ5yyse}
    }

  • Modeling Continuous Stochastic Processes with Dynamic Normalizing Flows
    Ruizhi Deng, Bo Chang, Marcus A. Brubaker, Greg Mori, Andreas Lehrmann
    Advances in Neural Information Processing Systems (NeurIPS), 2020.
    [arXiv] [Proceedings] [BibTeX]

    @inproceedings{deng2020modeling,
      title={Modeling Continuous Stochastic Processes with Dynamic Normalizing Flows},
      author={Deng, Ruizhi and Chang, Bo and Brubaker, Marcus A and Mori, Greg and Lehrmann, Andreas},
      booktitle={Advances in Neural Information Processing Systems},
      year={2020}
    }

  • Variational Hyper RNN for Sequence Modeling
    Ruizhi Deng, Yanshuai Cao, Bo Chang, Leonid Sigal, Greg Mori, Marcus A. Brubaker
    Preprint, 2020.
    [arXiv] [BibTeX]

    @article{deng2020variational,
      title={Variational Hyper {RNN} for Sequence Modeling},
      author={Deng, Ruizhi and Cao, Yanshuai and Chang, Bo and Sigal, Leonid and Mori, Greg and Brubaker, Marcus A},
      journal={arXiv preprint arXiv:2002.10501},
      year={2020}
    }

  • Point Process Flows
    Nazanin Mehrasa*, Ruizhi Deng*, Mohamed Osama Ahmed, Bo Chang, Jiawei He, Thibaut Durand, Marcus A. Brubaker, Greg Mori
    Temporal Point Processes (TPP) Workshop at NeurIPS, 2019.
    [arXiv] [BibTeX]

    @article{mehrasa2019point,
      title={Point Process Flows},
      author={Mehrasa, Nazanin and Deng, Ruizhi and Ahmed, Mohamed Osama and Chang, Bo and He, Jiawei and Durand, Thibaut and Brubaker, Marcus and Mori, Greg},
      journal={arXiv preprint arXiv:1910.08281},
      year={2019}.
    }

  • AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks
    Bo Chang, Minmin Chen, Eldad Haber, Ed H. Chi
    International Conference on Learning Representations (ICLR), 2019.
    [arXiv] [OpenReview] [BibTeX]

    @inproceedings{chang2019antisymmetric,
      title={Antisymmetric{RNN}: a dynamical system view on recurrent neural networks},
      author={Chang, Bo and Chen, Minmin and Haber, Eldad and Chi, Ed H},
      booktitle={International Conference on Learning Representations},
      year={2019},
      url={https://openreview.net/forum?id=ryxepo0cFX}
    }

  • Vine Copula Structure Learning via Monte Carlo Tree Search
    Bo Chang, Shenyi Pan, Harry Joe
    International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.
    [Proceedings] [GitHub] [BibTeX]

    @inproceedings{chang2019vine,
      title={Vine copula structure learning via {M}onte {C}arlo tree search},
      author={Chang, Bo and Pan, Shenyi and Joe, Harry},
      booktitle={International Conference on Artificial Intelligence and Statistics},
      year={2019}
    }

  • Dynamical Isometry and a Mean Field Theory of LSTMs and GRUs
    Dar Gilboa, Bo Chang, Minmin Chen, Greg Yang, Samuel S. Schoenholz, Ed H. Chi, Jeffrey Pennington
    Preprint, 2019.
    [arXiv] [BibTeX]

    @article{gilboa2019dynamical,
      title={Dynamical isometry and a mean field theory of {LSTM}s and {GRU}s},
      author={Gilboa, Dar and Chang, Bo and Chen, Minmin and Yang, Greg and Schoenholz, Samuel S and Chi, Ed H and Pennington, Jeffrey},
      journal={arXiv preprint arXiv:1901.08987},
      year={2019}
    }

  • Multi-level Residual Networks from Dynamical Systems View
    Bo Chang*, Lili Meng*, Eldad Haber, Frederick Tung, David Begert
    International Conference on Learning Representations (ICLR), 2018.
    [arXiv] [OpenReview] [BibTeX]

    @inproceedings{chang2018multilevel,
      title={Multi-level residual networks from dynamical systems view},
      author={Chang, Bo and Meng, Lili and Haber, Eldad and Tung, Frederick and Begert, David},
      booktitle={International Conference on Learning Representations},
      year={2018},
      url={https://openreview.net/forum?id=SyJS-OgR-}
    }

  • Reversible Architectures for Arbitrarily Deep Residual Neural Networks
    Bo Chang*, Lili Meng*, Eldad Haber, Lars Ruthotto, David Begert, Elliot Holtham
    AAAI Conference on Artificial Intelligence (AAAI), 2018.
    [arXiv] [Proceedings] [BibTeX]

    @inproceedings{chang2018reversible,
      title={Reversible architectures for arbitrarily deep residual neural networks},
      author={Chang, Bo and Meng, Lili and Haber, Eldad and Ruthotto, Lars and Begert, David and Holtham, Elliot},
      booktitle={AAAI Conference on Artificial Intelligence},
      year={2018}
    }

Computer Vision

  • Interpretable Spatio-Temporal Attention for Video Action Recognition
    Lili Meng, Bo Zhao, Bo Chang, Gao Huang, Wei Sun, Frederick Tung, Leonid Sigal
    IEEE International Conference on Computer Vision (ICCV) Workshops, 2019.
    [arXiv] [Proceedings] [BibTeX]

    @inproceedings{meng2018interpretable,
      title = {Interpretable Spatio-Temporal Attention for Video Action Recognition},
      author = {Meng, Lili and Zhao, Bo and Chang, Bo and Huang, Gao and Sun, Wei and Tung, Frederick and Sigal, Leonid},
      booktitle = {International Conference on Computer Vision (ICCV) Workshops},
      year={2019}
    }

  • Modular Generative Adversarial Networks
    Bo Zhao, Bo Chang, Zequn Jie, Leonid Sigal
    European Conference on Computer Vision (ECCV), 2018.
    [arXiv] [Proceedings] [Media Coverage (in Chinese)] [BibTeX]

    @inproceedings{zhao2018modular,
      title={Modular generative adversarial networks},
      author={Zhao, Bo and Chang, Bo and Jie, Zequn and Sigal, Leonid},
      booktitle={European Conference on Computer Vision},
      pages={150--165},
      year={2018}
    }

  • Generating Handwritten Chinese Characters Using CycleGAN
    Bo Chang*, Qiong Zhang*, Shenyi Pan, Lili Meng
    IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.
    [arXiv] [Proceedings] [GitHub] [BibTeX]

    @inproceedings{chang2018generating,
      title={Generating handwritten {C}hinese characters using {C}ycle{GAN}},
      author={Chang, Bo and Zhang, Qiong and Pan, Shenyi and Meng, Lili},
      booktitle={Winter Conference on Applications of Computer Vision},
      pages={199--207},
      year={2018}
    }

Statistics

  • Vine Regression with Bayes Nets: A Critical Comparison with Traditional Approaches Based on a Case Study on the Effects of Breastfeeding on IQ
    Roger M. Cooke, Harry Joe, Bo Chang
    Risk Analysis, 42(6), 1294–1305, 2022.
    [Journal] [BibTeX]

    @article{cooke2022vine,
      title={Vine Regression with Bayes Nets: A Critical Comparison with Traditional Approaches Based on a Case Study on the Effects of Breastfeeding on IQ},
      author={Cooke, Roger M and Joe, Harry and Chang, Bo},
      journal={Risk Analysis},
      volume = {42},
      number = {6},
      pages = {1294-1305},
      year = {2022},
      publisher={Wiley Online Library}
    }

  • Copula Diagnostics for Asymmetries and Conditional Dependence
    Bo Chang, Harry Joe
    Journal of Applied Statistics 47(9): 1587–1615, 2020.
    [Journal] [BibTeX]

    @article{chang2020copula,
      title = {Copula diagnostics for asymmetries and conditional dependence},
      author={Chang, Bo and Joe, Harry},
      journal={Journal of Applied Statistics},
      volume={47},
      number={9},
      pages={1587--1615},
      year={2020},
      publisher={Taylor & Francis}
    }

  • Vine Copula Regression for Observational Studies
    Roger M. Cooke, Harry Joe, Bo Chang
    AStA Advances in Statistical Analysis 104: 141–167, 2020.
    [Journal] [BibTeX]

    @article{cooke2020vine,
      title={Vine copula regression for observational studies},
      author={Cooke, Roger M and Joe, Harry and Chang, Bo},
      journal={AStA Advances in Statistical Analysis},
      volume={104},
      pages={141--167},
      year={2020},
      publisher={Springer}
    }

  • Prediction Based on Conditional Distributions of Vine Copulas
    Bo Chang, Harry Joe
    Computational Statistics & Data Analysis 139: 45–63, 2019.
    [arXiv] [Journal] [BibTeX]

    @article{chang2019prediction,
      title={Prediction based on conditional distributions of vine copulas},
      author={Chang, Bo and Joe, Harry},
      journal={Computational Statistics \& Data Analysis},
      volume={139},
      pages={45--63},
      year={2019},
      publisher={Elsevier}
    }

  • Vine Copulas: Dependence Structure Learning, Diagnostics, and Applications to Regression Analysis
    Bo Chang
    Ph.D. Thesis, University of British Columbia, 2019.
    [UBC Library] [BibTeX]

    @phdthesis{chang2019vinecopulas,
      series={Electronic Theses and Dissertations (ETDs) 2008+},
      title={Vine copulas: dependence structure learning, diagnostics, and applications to regression analysis},
      url={https://open.library.ubc.ca/collections/ubctheses/24/items/1.0379699},
      DOI={http://dx.doi.org/10.14288/1.0379699},
      school={University of British Columbia},
      author={Chang, Bo},
      year={2019},
      collection={Electronic Theses and Dissertations (ETDs) 2008+}
    }

  • Vine Regression
    Roger M. Cooke, Harry Joe, Bo Chang
    Resources for the Future Discussion Paper 15-52, 2015.
    [SSRN] [BibTeX]

    @article{cooke2015vine,
      title={Vine regression},
      author={Cooke, Roger M and Joe, Harry and Chang, Bo},
      journal={Resources for the Future Discussion Paper},
      volume={15-52},
      year={2015}
    }

* Equal Contribution