van der Schaar Lab

van der Schaar Lab at NeurIPS 2021: 14 papers and 3 workshops

Note: this post originally appeared on September 30, but was updated and republished on December 2 with details regarding poster sessions and workshops.

The van der Schaar Lab will make another strong showing at NeurIPS—widely considered the world’s largest and most prestigious AI and machine learning research conference—with Mihaela van der Schaar giving 2 invited talks at workshops, and a record total of 14 papers accepted for publication this year.

This year’s NeurIPS conference will run from December 6 through 14.

Details regarding both workshops all all 14 papers (including timing of poster sessions, etc.) can be found below.

Workshops – December 13 and 14

On December 13 at 13:30 GMT, Mihaela van der Schaar will give an invited talk at the 2021 Workshop on Meta-Learning (MetaLearn 2021). Her talk will introduce quantitative epistemology, a transformational new area of research pioneered by our lab as a strand of machine learning aimed at understanding, supporting, and improving human decision-making.

On December 14 at 21:00 GMT, Mihaela van der Schaar will deliver an invited talk on the topic of Self-supervised Learning for Genomics at the 2021 Self-supervised Learning Workshop.

On December 14, Mihaela will address the Deep Generative Models and Downstream Applications Workshop on the topic of Synthetic Data Generation and Assessment: Challenges, Methods, Impact. This is one of our lab’s core research pillars, as we believe synthetic data has the potential to revolutionize how we access and interact with datasets in and beyond healthcare.

14 papers – details and poster times

This result represents the most papers accepted for the lab at any conference to date, and perfectly captures the diverse strengths of its small research team. The papers cover a number of the lab’s key research pillars, such as interpretable machine learning, individualized treatment effect inference, understanding and supporting decision-making (quantitative epistemology), and time series analysis, among others.

Titles, authors and abstracts for all 14 accepted papers are given below.

SurvITE: Learning Heterogeneous Treatment Effects from Time-to-Event Data

Alicia Curth, Changhee Lee, Mihaela van der Schaar

Abstract

On Inductive Biases for Heterogeneous Treatment Effect Estimation
(spotlight paper)

Alicia Curth, Mihaela van der Schaar

Abstract and paper URL

Really Doing Great at Estimating CATE? A Critical Look at ML Benchmarking Practices in Treatment Effect Estimation

Alicia Curth, David Svensson, Jim Weatherall Mihaela van der Schaar

Abstract and paper URL Abstract

Estimating Multi-cause Treatment Effects via Single-cause Perturbation

Zhaozhi Qian, Alicia Curth, Mihaela van der Schaar

Abstract and paper URL

Integrating Expert ODEs into Neural ODEs: Pharmacology and Disease Progression

Zhaozhi Qian, William R. Zame, Lucas Fleuren, Paul Elbers, Mihaela van der Schaar

Abstract and paper URL

SyncTwin: Treatment Effect Estimation with Longitudinal Outcomes

Zhaozhi Qian, Yao Zhang, Ioana Bica, Angela Wood, Mihaela van der Schaar

Abstract and paper URL

Invariant Causal Imitation Learning for Generalizable Policies

Ioana Bica, Daniel Jarrett, Mihaela van der Schaar

Abstract and paper URL

Time-series Generation by Contrastive Imitation

Daniel Jarrett, Ioana Bica, Mihaela van der Schaar

Abstract and paper URL

The Medkit-Learn(ing) Environment: Medical Decision Modelling through Simulation

Alex Chan, Ioana Bica, Alihan Hüyük, Daniel Jarrett, Mihaela van der Schaar

Abstract and paper URL

Closing the loop in medical decision support by understanding clinical decision-making: A case study on organ transplantation

Yuchao Qin, Fergus Imrie, Alihan Hüyük, Daniel Jarrett, Alexander Gimson, Mihaela van der Schaar

Abstract and paper URL

Explaining Latent Representations with a Corpus of Examples
(spotlight paper)

Jonathan Crabbé, Zhaozhi Qian, Fergus Imrie, Mihaela van der Schaar

Abstract and paper URL

Abstract and paper URL

MIRACLE: Causally-Aware Imputation via Learning Missing Data Mechanisms

Trent Kyono, Yao Zhang, Alexis Bellot, Mihaela van der Schaar

Abstract and paper URL

DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networks

Trent Kyono, Boris van Breugel, Jeroen Berrevoets, Mihaela van der Schaar

Abstract and paper URL

About NeurIPS

For a full list of the van der Schaar Lab’s publications, click here.

Nick Maxfield

From 2020 to 2022, Nick oversaw the van der Schaar Lab’s communications, including media relations, content creation, and maintenance of the lab’s online presence.