van der Schaar Lab

van der Schaar Lab at ICML 2021: tutorial, 4 papers, and 4 workshops

Note: this post originally appeared on May 10, but was updated and republished on July 14 with details regarding the ICML 2021 tutorial, poster sessions and workshops.

The van der Schaar Lab’s work will be highly visible at ICML 2021 (July 18-24), the leading international academic conference in machine learning. Mihaela van der Schaar and postdoc Ahmed Alaa will deliver a tutorial on synthetic data, while the lab’s researchers will publish 4 papers and deliver talks at 3 separate workshops being held as part of the conference—as well as co-organizing a workshop.

Along with NeurIPS and ICLR, ICML is one of the 3 primary conferences of high impact in machine learning and artificial intelligence research.

Tutorial on synthetic data (July 19)

On July 19 at 17:00 CEST (16:00 BST; other time zones here), Mihaela van der Schaar and postdoc Ahmed Alaa will deliver a tutorial entitled “Synthetic healthcare data generation and assessment: challenges, methods, and impact on machine learning.”

Covering both high-level theory and specific methods and approaches, the tutorial will advance and explain a vision for synthetic data to help catalyze a revolution in healthcare by breaking the current logjam in data availability. In addition, Mihaela and Ahmed will discuss the issue of evaluating the quality of synthetic data and the performance of generative models; they will highlight the challenges associated with evaluating generative models as compared to discriminative predictions, and present various metrics that can be used to quantify different aspects of synthetic data quality.

Papers accepted for publication

All 4 papers accepted for publication at ICML 2021 represent cutting-edge machine learning methods on complex and important problems, including understanding human decision-making, individualized treatment effects, time series analysis, and machine learning interpretability. All make substantial technical contributions in areas the van der Schaar Lab believes to be particularly promising, and which further the lab’s research agenda for healthcare.

Titles, authors and abstracts for all 4 papers are given below.

Inverse Decision Modeling: Learning Interpretable Representations of Behavior

Daniel Jarrett, Alihan Hüyük, Mihaela van der Schaar

Abstract

Explaining Time Series Predictions With Dynamic Masks

Jonathan Crabbé, Mihaela van der Schaar

Abstract

Policy Analysis using Synthetic Controls in Continuous-Time

Alexis Bellot, Mihaela van der Schaar

Abstract

Learning Queueing Policies for Organ Transplantation Allocation using Interpretable Counterfactual Survival Analysis

Jeroen Berrevoets, Ahmed Alaa, Zhaozhi Qian, James Jordon, Alexander Gimson, Mihaela van der Schaar

Abstract

Keynote at Interpretable Machine Learning for Healthcare Workshop (July 23)

On July 23 at 06:30 PDT (14:30 BST; other time zones here), Mihaela van der Schaar will give a keynote as part of the Interpretable Machine Learning in Healthcare (IMLH) Workshop.

The talk, entitled “Quantitative epistemology: conceiving a new human-machine partnership,” will introduce and explain a brand new area of research pioneered by the van der Schaar Lab with the aim of using AI and machine learning to understand and empower human learning and decision-making.

Keynote at Time Series Workshop (July 24)

On July 24 at 09:00 PDT (17:00 BST; other time zones here), Mihaela van der Schaar will deliver a keynote as part of the Time Series Workshop.

Her talk will be entitled “Time-series in healthcare: challenges and solutions,” and will explore new approaches to building dynamic models that incorporate time series datasets available in healthcare.

Presentation of paper at Neglected Assumptions in Causal Inference Workshop (July 23)

On July 23, Ph.D. student Alicia Curth will present a paper entitled “Doing Great at Estimating CATE? On the Neglected Assumptions in Benchmark Comparisons of Treatment Effect Estimators” as part of the Workshop on The Neglected Assumptions in Causal Inference (NACI).

Self-Supervised Learning for Reasoning and Perception Workshop (July 24)

Mihaela van der Schaar is on the organizing team for the Self-Supervised Learning for Reasoning and Perception Workshop, which will take place on July 24. The workshop will bring together SSL-interested researchers from various domains to discuss how to develop SSL methods for reasoning tasks, such as how to design pretext tasks for symbolic reasoning, how to develop contrastive learning methods for relational reasoning, how to develop SSL approaches to bridge reasoning and perception, etc.

About ICML 2021


The full ICML 2021 schedule will be available here.

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.