Neuro-symbolic Neurodegenerative Disease Modeling as Probabilistic Programmed Deep Kernels
release_snzegvlnyzdn3fibg56wsurbki
by
Alexander Lavin
2021
Abstract
We present a probabilistic programmed deep kernel learning approach to
personalized, predictive modeling of neurodegenerative diseases. Our analysis
considers a spectrum of neural and symbolic machine learning approaches, which
we assess for predictive performance and important medical AI properties such
as interpretability, uncertainty reasoning, data-efficiency, and leveraging
domain knowledge. Our Bayesian approach combines the flexibility of Gaussian
processes with the structural power of neural networks to model biomarker
progressions, without needing clinical labels for training. We run evaluations
on the problem of Alzheimer's disease prediction, yielding results that surpass
deep learning in both accuracy and timeliness of predicting neurodegeneration,
and with the practical advantages of Bayesian nonparametrics and probabilistic
programming.
In text/plain
format
Archived Files and Locations
application/pdf 666.2 kB
file_5dwszqsrjnfxthlwmfcderhh2u
|
arxiv.org (repository) web.archive.org (webarchive) |
2009.07738v3
access all versions, variants, and formats of this works (eg, pre-prints)