Elliot J. Crowley
Lecturer, University of Edinburgh
group · github · scholar

I am a Lecturer (Assistant Professor) in Machine Learning and Computer Vision at the School of Engineering in the University of Edinburgh. I run the Bayesian and Neural Systems research group with Amos Storkey from the School of Informatics.

My research interests include:

  • simplifying deep learning
  • neural architecture search
  • efficient training and inference
  • weakly supervised learning in vision
  • applying machine learning to engineering

Before, I was a postdoc at the School of Informatics in Edinburgh. I did my MEng and DPhil at the University of Oxford. For my DPhil, I worked with Andrew Zisserman on object recognition and retrieval in art.

Selected Publications

Prediction-Guided Distillation for Dense Object Detection
C. Yang, M. Ochal, A. Storkey, E. J. Crowley
[pdf] [arXiv] [code]

Contrastive Object-level Pre-training with Spatial Noise Curriculum Learning
C. Yang, L. Huang, E. J. Crowley
[pdf] [arXiv] [code]

Neural Architecture Search without Training
J. Mellor, J. Turner, A. Storkey, E. J. Crowley
ICML 2021 (Long talk)
[pdf] [arXiv] [code]

Neural Architecture Search as Program Transformation Exploration
J. Turner, E. J. Crowley, M. O’Boyle
ASPLOS 2021 (Distinguished Paper)
[pdf] [arXiv] [code]

Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels
M. Patacchiola, J. Turner, E. J. Crowley, M. O’Boyle, A. Storkey
NeurIPS 2020 (Spotlight)
[pdf] [arXiv] [code]

BlockSwap: Fisher-guided Block Substitution for Network Compression on a Budget
J. Turner*, E. J. Crowley*, M. O’Boyle, A. Storkey, G. Gray
ICLR 2020
[pdf] [arXiv] [code]

* indicates equal contribution. Full list of publications is on scholar.
Thanks to Jack Turner for the website template.