Our weekly SRI Seminar Series welcomes Chris Maddison, an assistant professor in the Department of Computer Science and the Department of Statistical Sciences at the University of Toronto. In addition to his role as a faculty affiliate at the Schwartz Reisman Institute, Maddison is also a CIFAR AI Chair at the Vector Institute, a research scientist at DeepMind, and a member of the ELLIS Society. Maddison was recently named the 2021 recipient of the Corcoran Memorial Prize at the University of Oxford.
Maddison works on the methodology of statistical machine learning, with an emphasis on methods that work at scale in deep learning applications. His research interests lie in the study of Bayesian inference, optimization, discrete search. Maddison was a founding member of the AlphaGo project, which received the IJCAI Marvin Minsky Medal for Outstanding Achievements in AI in 2018.
Talk title:
“The future of representation learning”
Abstract:
A trend is emerging in machine learning: well-resourced firms are training massive models on vast datasets and making those models available for adaptation and use by users on downstream tasks. For instance, OpenAI's CLIP model, trained on 400 million image and captions, is able to learn powerful visual representations that are now being used in many downstream applications, such as AI-generated art. This centralization creates risk: the provenance of the data is often unknown, which makes it difficult to assess whether pipelines that incorporate these representations are fair or robust. Centralizing representation learning also holds promise. It can help us manage the data and energy resources required by downstream applications, as well as democratize the expertise and computing resources of massive firms. In this talk, I will walk us through recent developments in representation learning and speculate on the future that these developments foreshadow.
Suggested readings:
OpenAI, “CLIP: Connecting Text and Images.” January 5, 2021.
A. Radford et al., “Learning transferable visual models from natural language supervision.” 2021.
R. Bommasani et al., “On the opportunities and risks of foundation models.” 2021.
Y. Bengio, A. Courville, P. Vincent, “Representation learning: A review and new perspectives.” IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8): 1798-1828. August 2013.
About Chris Maddison
Chris Maddison is an assistant professor in the Department of Computer Science and the Department of Statistical Sciences at the University of Toronto. He is a CIFAR AI Chair at the Vector Institute, a research scientist at DeepMind, a member of the ELLIS Society, and a faculty affiliate of the Schwartz Reisman Institute for Technology and Society. Maddison works on the methodology of statistical machine learning, with an emphasis on methods that work at scale in deep learning applications. His research interests lie in the study of Bayesian inference, optimization, discrete search. Previously, he was a member at the Institute for Advanced Study in Princeton, NJ from 2019–2020, and he completed his DPhil at the University of Oxford. Maddison was an Open Philanthropy AI Fellow during his graduate studies. He received a NeurIPS Best Paper Award in 2014. He was a founding member of the AlphaGo project, which received the IJCAI Marvin Minsky Medal for Outstanding Achievements in AI in 2018.
About the SRI Seminar Series
The SRI Seminar Series brings together the Schwartz Reisman community and beyond for a robust exchange of ideas that advance scholarship at the intersection of technology and society. Seminars are led by a leading or emerging scholar and feature extensive discussion.
Each week, a featured speaker will present for 45 minutes, followed by 45 minutes of discussion. Registered attendees will be emailed a Zoom link approximately one hour before the event begins. The event will be recorded and posted online.
Chris Maddison