Van der Schaar Lab Sets Record with 14 Papers at NeurIPS 2021
The van der Schaar Lab achieves a record-breaking presence at NeurIPS 2021 with 14 accepted papers and three workshop presentations led by Mihaela van der Schaar.
The van der Schaar Lab makes a major impact at NeurIPS 2021, widely considered the world's most prestigious AI and machine learning conference, by securing a record total of 14 accepted papers. This impressive milestone represents the highest number of papers the lab has ever had accepted at a single conference, highlighting the diverse strengths of its dedicated research team.
Mihaela van der Schaar delivers three invited talks during the event's workshop days on December 13 and 14. She introduces quantitative epistemology at the MetaLearn 2021 workshop, discusses self-supervised learning for genomics at the Self-supervised Learning Workshop, and addresses synthetic data generation and assessment at the Deep Generative Models Workshop.
The 14 accepted papers cover several of the lab's core research pillars, including interpretable machine learning, individualized treatment effect inference, and time series analysis. One highlighted paper, SurvITE by Alicia Curth, Changhee Lee, and Mihaela van der Schaar, focuses on learning heterogeneous treatment effects from time-to-event data using deep learning and causal inference.