Search

Graham Taylor

LMB_GrahamTaylor

Appointment

  • CIFAR Azrieli Global Scholar 2016
  • Learning in Machines & Brains

Institution

  • University of Guelph
  • Vector Institute
School of Engineering

Country

  • Canada

Education

PhD (Machine Learning), University of Toronto
MAS (Systems Design Engineering), University of Waterloo
BAS (Systems Design Engineering), University of Waterloo

About

Graham Taylor is a machine learning researcher. He seeks to discover new algorithms and architectures for deep learning: the automatic construction of hierarchical algorithms from high-dimensional, unstructured data.

He is especially interested in time series, having applied his work to better understand human and animal behaviour, environmental data (climate or agricultural), audio (music or speech) and financial time series. His work also intersects high performance computing, investigating better ways to leverage hardware accelerators to cope with the challenges of large-scale machine learning.

Taylor is active in promoting entrepreneurial activities in artificial intelligence. He is the academic director of NextAI, a non-profit initiative to establish Canada as the AI hub for research, venture creation and technology commercialization.

Awards

Canada Research Chair in Machine Learning (effective September 2018)

Relevant Publications

Devries, T., and G.W. Taylor. “Dataset Augmentation in Feature Space.” In International Conference on Learning Representations (ICLR) Workshop Track, 2017.

Neverova, N. et al. “ModDrop: Adaptive Multi-Modal Gesture Recognition.” TPAMI 38, no. 8 (2016): 1692–1706.

Neverova, N. et al. “Learning Human Identity From Motion Patterns.” IEEE Access 4 (2016): 1810–1820.

Zeiler, M.D., G.W. Taylor, and R. Fergus. 2011. “Adaptive Deconvolutional Networks for Mid and High Level Feature Learning.” In Proceedings of 2011 International Conference on Computer Vision, 2018–25.

Taylor, G.W., G.E. Hinton, and S.T. Roweis. “Two Distributed-State Models For Generating High-Dimensional Time Series.” Journal of Machine Learning Research: JMLR 12 (March 2011): 1025–1068.

Connect

Website