Skip to content
Created with GIMP

Yuhong Guo

Appointment

  • Canada CIFAR AI Chair
  • Pan-Canadian AI Strategy

Connect

Google Scholar

About

Yuhong Guo’s research seeks to expand the autonomy of machine learning methods by reducing the dependence on extensive human guidance. Her research program has been founded on three main directions: generalized transfer learning, learning from incomplete data, and learning from weakly supervised data. Yuhong’s work on transfer learning has leveraged auxiliary data resources to reduce the annotation requirement for target tasks, via domain adaptation,zero-shot and few-shot learning. For data analysis scenarios,such as recommender systems, where the data is naturally sparse and contains missing entries, she has developed matrix completion techniques to infer the underlying data association mechanisms and automatically recover missing observations. Moreover, she has developed techniques that can learn accurate prediction models from weak supervision such as imprecise annotations collected via crowdsourcing from non-experts. This research program has broadened the applicability of machine learning to more diverse scenarios.

Awards

  • Canada Research Chair Tier 2, NSERC, 2016
  • Outstanding Paper Award, AAAI, 2012
  • NSERC Postdoctoral Fellowship, NSERC, 2008
  • Distinguished Paper Award, IJCAI, 2005

Relevant Publications

  • Zhao, Z., Guo, Y., Shen, H., & Ye, J. (2020). Adaptive Object Detection with Dual Multi-Label Prediction. arXiv preprint arXiv:2003.12943.
  • Ye, M., & Guo, Y. (2019). Progressive ensemble networks for zero-shot recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 11728-11736).
  • Guo, Y. (2017, February). Convex co-embedding for matrix completion with predictive side information. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 31, No. 1).
  • Xiao, M., & Guo, Y. (2014). Feature space independent semi-supervised domain adaptation via kernel matching. IEEE transactions on pattern analysis and machine intelligence, 37(1), 54-66.
  • Xiao, M., & Guo, Y. (2013). A novel two-step method for cross language representation learning. Advances in Neural Information Processing Systems, 26, 1259-1267.

Support Us

CIFAR is a registered charitable organization supported by the governments of Canada, Alberta and Quebec, as well as foundations, individuals, corporations and Canadian and international partner organizations.

MaRS Centre, West Tower
661 University Ave., Suite 505
Toronto, ON M5G 1M1 Canada