Compositional modeling of nonlinear dynamical systems with ode-based random features

T McDonald, M Álvarez - Advances in neural information …, 2021 - proceedings.neurips.cc
Effectively modeling phenomena present in highly nonlinear dynamical systems whilst also
accurately quantifying uncertainty is a challenging task, which often requires problem …

Joint models for event prediction from time series and survival data

X Yue, RA Kontar - Technometrics, 2021 - Taylor & Francis
We present a nonparametric prognostic framework for individualized event prediction based
on joint modeling of both time series and time-to-event data. Our approach exploits a …

Adjoint-aided inference of Gaussian process driven differential equations

P Gahungu, C Lanyon, MA Álvarez… - Advances in …, 2022 - proceedings.neurips.cc
Linear systems occur throughout engineering and the sciences, most notably as differential
equations. In many cases the forcing function for the system is unknown, and interest lies in …

Deep Latent Force Models: ODE-based Process Convolutions for Bayesian Deep Learning

T Baldwin-McDonald, MA Álvarez - arXiv preprint arXiv:2311.14828, 2023 - arxiv.org
Effectively modeling phenomena present in highly nonlinear dynamical systems whilst also
accurately quantifying uncertainty is a challenging task, which often requires problem …

Learning nonparametric Volterra kernels with Gaussian processes

M Ross, MT Smith, M Álvarez - Advances in neural …, 2021 - proceedings.neurips.cc
This paper introduces a method for the nonparametric Bayesian learning of nonlinear
operators, through the use of the Volterra series with kernels represented using Gaussian …

Physically-inspired Gaussian process models for post-transcriptional regulation in Drosophila

AF López-Lopera, N Durrande… - IEEE/ACM transactions …, 2019 - ieeexplore.ieee.org
The regulatory process of Drosophila is thoroughly studied for understanding a great variety
of biological principles. While pattern-forming gene networks are analyzed in the …

Multioutput convolution spectral mixture for Gaussian processes

K Chen, T van Laarhoven, P Groot… - … on Neural Networks …, 2019 - ieeexplore.ieee.org
Multioutput Gaussian processes (MOGPs) are an extension of Gaussian processes (GPs) for
predicting multiple output variables (also called channels/tasks) simultaneously. In this …

Generalized convolution spectral mixture for multitask Gaussian processes

K Chen, T van Laarhoven, P Groot… - IEEE transactions on …, 2020 - ieeexplore.ieee.org
Multitask Gaussian processes (MTGPs) are a powerful approach for modeling
dependencies between multiple related tasks or functions for joint regression. Current …

Adaptive RKHS Fourier Features for Compositional Gaussian Process Models

X Shi, T Baldwin-McDonald, MA Álvarez - arXiv preprint arXiv:2407.01856, 2024 - arxiv.org
Deep Gaussian Processes (DGPs) leverage a compositional structure to model non-
stationary processes. DGPs typically rely on local inducing point approximations across …

Patient-specific effects of medication using latent force models with Gaussian processes

LF Cheng, B Dumitrascu, M Zhang… - International …, 2020 - proceedings.mlr.press
A multi-output Gaussian process (GP) is a flexible Bayesian nonparametric framework that
has proven useful in jointly modeling the physiological states of patients in medical time …