Inspiration Exchange: returning on October 7

Following a short break over the summer, the van der Schaar Lab’s Inspiration Exchange engagement series for machine learning students and professionals will resume on October 7, with a session on time series for healthcare.

Inspiration Exchange: looking back at the last year

Roughly a year ago, Mihaela van der Schaar started the Inspiration Exchange engagement series for machine learning students and professionals. The aim of the series was to discuss machine learning models and techniques, share ideas about how machine learning can revolutionize healthcare, spark new projects and collaborations, and raise awareness about healthcare as a unique and exciting area of machine learning.

The 10 sessions held so far have explored a wide range of topics, with many more lined up.

Links to all 10 sessions are provided below:
Session 1: introduction to automated machine learning
Session 2: automated machine learning pipelines
Session 3: software packages for automated machine learning
Session 4: recent projects in machine learning for healthcare
Session 5: synthetic data concepts and approaches
Session 6: synthetic data evaluation
Session 7: application-oriented projects in machine learning for healthcare
Session 8: individualized treatment effect inference (key concepts and challenges)
Session 9: individualized treatment effect inference (time series)
Session 10: quantitative epistemology

October 7 session on time series for healthcare

The theme of the upcoming October 7 Inspiration Exchange session will be time series for healthcare—an absolutely vital topic that receives insufficient consideration.

The transformation of healthcare through machine learning depends heavily on the successful application of time series data to model longitudinal trajectories for health and disease, but the development of models for time series is a complex, hard-to-define research task that touches every other area of machine learning for healthcare. Additionally, there are some model development challenges that are unique to the time series setting.

This topic has been explored extensively in a recent write-up.

The upcoming session will examine the role of time series at the heart of machine learning for healthcare, and discuss the many complex challenges specific to this setting. Mihaela and members of the lab will introduce a few important new approaches and methods, and wrap up with a participative Q&A.