Inspiration Exchange #28

We’ll be holding our next Inspiration Exchange session on May 23 at 16:00 BST.

We are glad to have met many of you at AISTATS and ICLR over the past two weeks! On Tuesday, May 23 at 4 PM BST (other timezones here), we will continue with presentations of the works our lab showcased at AISTATS’23 and ICLR’23 encompassing treatment effect estimation, synthetic data, neural network regularization, and temporal clustering! 

In the upcoming session, we will have:

Alicia Curth presents her work on inferring heterogeneous treatment effects from time-to-event data in the presence of competing risks;
Boris van Breugel presents DOMIAS, a density-based membership inference attack model that aims to infer membership from synthetic data by detecting local overfitting of generative models;
Alan Jeffares presents TANGOS, a novel regularization method built on latent neuron attributions that encourages orthogonalization and specialization of neuron attributions to improve generalization performance;
Yuchao Qin presents T-Phenotype, a novel temporal clustering method that discovers phenotypes of predictive temporal patterns from time-series data. 

We encourage you to learn more about the presented work by reading the described papers. The titles, authors, abstracts, and perceived impacts of all 11 accepted papers can be found on our lab website. If you missed the previous session, which presented 4 additional accepted works, we encourage you to watch the recordings on YouTube!