NorwAI&NAIL AI Seminar «Learning to X» with Prof. Kyunghyun Cho

About the speaker

Prof. Kyunghyun Cho is a leading expert in machine learning and natural language processing, serving as a professor of computer science and data science at New York University (NYU) and as the Co-Founder and Senior Director of Frontier Research at Genentech’s Prescient Design. He is renowned as the inventor of the Gated Recurrent Unit (GRU), a foundational architecture in recurrent neural networks widely used in AI. Prof. Cho has held prominent roles in the AI community, including Program Chair for ICLR, NeurIPS, and ICML, and a founding co-Editor-in-Chief of the Transactions on Machine Learning Research. He is also a CIFARFellow and an Associate Member of the National Academy of Engineering of Korea. He received the Samsung Ho-Am Prize in Engineering in 2021 for his distinguished work on neural machine translation. Prof. Cho continues to shape AI research and applications, driving innovation at the intersection of academia and industry.

Abstract

In the first part of this talk, Prof. Cho will talk about the idea of meta-learning by going over a few representative earlier studies, including neural processes,learned optimizers, hyperparameter tuning and in-context learning, in order to motivate audience to think of neural net learning from a broader perspective. Hewill then talk about how this perspective allows us to think of neural net learning as a way to learn an algorithm, enabling us to arrive at algorithms we could not have ourselves. In the second part of his talk, he will describe two ongoing studies on causal discovery and causal inference.

Some relevant papers for the talk

• https://proceedings.mlr.press/v70/finn17a.html
• https://arxiv.org/abs/1807.01622
• https://arxiv.org/abs/1606.04474
• https://arxiv.org/abs/2404.11018
• https://arxiv.org/abs/2408.16218
• https://arxiv.org/abs/2102.07813
• https://www.nature.com/articles/s41586-023-06668-3

Details

Time: Tuesday, January 14th, 10:00-11:00

Place: IT Building South, 1st floor, Room 119 Drivhuset, Gløshaugen (Entrance at Sem Sælandsv. 7)

Competency level of the talks: Technical

Online participation: Use this link to participate online in Teams. The seminar series is mainly aimed for in person participation, but you are welcome to join the Teams stream.