
David Rolnick
Appointment
Canada CIFAR AI Chair
National Program Committee member
Pan-Canadian AI Strategy
About
David Rolnick is a Canada CIFAR AI Chair at Mila and an assistant professor in the School of Computer Science at McGill University. He is the co-founder and chair of Climate Change AI.
Rolnick studies machine learning innovations for tackling climate change, as well as the mathematical foundations of deep learning algorithms. His work bridges the gap between machine learning research and fields such as energy and ecology via high-impact applications in climate change mitigation and adaptation. This work has significantly influenced international policy and entrepreneurship.
Rolnick’s work on the foundations of deep learning includes analysis of when and why different algorithms succeed or fail, as well as the development of new algorithms leveraging insights from mathematics and computational neuroscience. This work aims to build the groundwork for next-generation deep learning research by replacing trial and error progress with algorithms designed from first principles.
Awards
- MIT Technology Review, “35 Innovators Under 35” Global List, 2021
- NSF Mathematical Sciences Postdoctoral Research Fellow, 2018-2020
- Hugh Hampton Young Memorial Fund Fellowship, 2017-2018
- NSF Graduate Research Fellowship, 2013-2017
- US-Germany Fulbright Study/ Research Scholarship, 2012-2013
Relevant Publications
Rolnick, D., & Kording, K. (2020). Reverse-engineering deep ReLU networks. In International Conference on Machine Learning (pp. 8178-8187). PMLR.
Rolnick, D., Donti, P. L., Kaack, L. H., Kochanski, K., Lacoste, A., Sankaran, K., Bengio, Y. & et. al. (2019). Tackling climate change with machine learning.
Rolnick, D., Ahuja, A., Schwarz, J., Lillicrap, T., & Wayne, G. (2019). Experience replay for continual learning. In Advances in Neural Information Processing Systems (pp. 350-360).
Hanin, B., & Rolnick, D. (2019). Complexity of linear regions in deep networks.
Rolnick, D., & Tegmark, M. (2017). The power of deeper networks for expressing natural functions.
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.