5 papers
Constraining Variational Inference with Geometric Jensen-Shannon Divergence
Jacob Deasy, Nikola Simidjievski, Pietro Liò
We examine the problem of controlling divergences for latent space regularisation in variational autoencoders. Specifically, when aiming to reconstruct example …
Adaptive Prediction Timing for Electronic Health Records
Jacob Deasy, Ari Ercole, Pietro Liò
In realistic scenarios, multivariate timeseries evolve over case-by-case time-scales. This is particularly clear in medicine, where the rate of clinical events varies by ward, pati…
The r-Hunter-Saxton equation, smooth and singular solutions and their approximation
Colin Cotter, Jacob Deasy, Tristan Pryer
In this work we introduce the r-Hunter-Saxton equation, a generalisation of the Hunter-Saxton equation arising as extremals of an action principle posed in L_r. We characterise sol…
Impact of novel aggregation methods for flexible, time-sensitive EHR prediction without variable selection or cleaning
Jacob Deasy, Ari Ercole, Pietro Liò
Dynamic assessment of patient status (e.g. by an automated, continuously updated assessment of outcome) in the Intensive Care Unit (ICU) is of paramount importance for early alerti…
Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or pre-processing
Jacob Deasy, Pietro Liò, Ari Ercole
We present a machine learning pipeline and model that uses the entire uncurated EHR for prediction of in-hospital mortality at arbitrary time intervals, using all available chart,…