3:30pm - 4:30pm | Biostat Seminar: Reframing proportional-hazards modeling for large time-to-event datasets with applications to deep learning

Date: 
Wednesday, December 2, 2020

Noah Simon
Associate Professor
Department of Biostatistics
University of Washington

Wednesday, December 02, 2020
3:30pm – 4:40pm, via Zoom
https://ucla.zoom.us/j/98576333860?pwd=QTdSdmVZOWMwaHZscldJZG1GUzhBQT09
Meeting ID: 985 7633 3860
Passcode: 140409

To build inferential or predictive survival models, it is common to assume
proportionality of hazards and fit a model by maximizing the partial likelihood. This
has been combined with non-parametric and high dimensional techniques, eg.
spline expansions and penalties, to flexibly build survival models.
New challenges require extension and modification of that approach. In a number
of modern applications there is interest in using complex features such as images
to predict survival. In these cases, it is necessary to connect more modern
backends to the partial likelihood (such as deep learning infrastructures based on
eg. convolutional/recurrent neural networks). In such scenarios, large numbers of
observations are needed to train the model. However, in cases where those
observations are available, the structure of the partial likelihood makes
optimization difficult (if not completely intractable).
In this talk we show how the partial likelihood can be simply modified to easily deal
with large amounts of data. In particular, with this modification, stochastic gradient-
based methods, commonly applied in deep learning, are simple to employ. This
simplicity holds even in the presence of left truncation/right censoring. This can
also be applied relatively simply with data stored in a distributed manner.