Mengyang Gu, UC Santa Barbara
School of Statistics Seminar
Event Date & Time
|
-
Event Location
150 Ford Hall
224 Church St SE
Minneapolis,
MN
55455
Mengyang Gu, UC Santa Barbara
School of Statistics Seminar
GNet: A scalable and flexible Gaussian process network accelerated by the jointly inverse Kalman filter
ABSTRACT
We build GNet, a scalable and flexible Gaussian process network with nonparametric activation functions. To reduce computational and storage costs, we develop the jointly inverse Kalman filter, a fast algorithm together with closed-form expressions of gradients for accelerating model training and predictions without the need to form covariance matrices. Using a unified optimization setting, GNet shows competitive performance across a diverse range of test problems, including predicting nonlinear functions, nonparametric regression of real-world data, and predicting one-body direct correlation functions with high-dimensional inputs in classical density functional theory. The strong performance of GNet, accelerated by the jointly inverse Kalman filter, suggests broad applicability to large-scale predictive modeling with substantially reduced computational and storage costs.
BIO
Mengyang Gu is an associate professor in the Department of Statistics and Applied Probability at UC Santa Barbara. He obtained a PhD in Statistical Science from Duke University in 2016. He has expertise in Bayesian analysis and uncertainty quantification. He focuses on developing scalable, accurate and flexible predictive models, and inverse estimation approaches with applications in physics, materials science and engineering. He received the SIAM activity group on uncertainty quantification (SIAG/UQ) early career prize in 2022.
We build GNet, a scalable and flexible Gaussian process network with nonparametric activation functions. To reduce computational and storage costs, we develop the jointly inverse Kalman filter, a fast algorithm together with closed-form expressions of gradients for accelerating model training and predictions without the need to form covariance matrices. Using a unified optimization setting, GNet shows competitive performance across a diverse range of test problems, including predicting nonlinear functions, nonparametric regression of real-world data, and predicting one-body direct correlation functions with high-dimensional inputs in classical density functional theory. The strong performance of GNet, accelerated by the jointly inverse Kalman filter, suggests broad applicability to large-scale predictive modeling with substantially reduced computational and storage costs.
BIO
Mengyang Gu is an associate professor in the Department of Statistics and Applied Probability at UC Santa Barbara. He obtained a PhD in Statistical Science from Duke University in 2016. He has expertise in Bayesian analysis and uncertainty quantification. He focuses on developing scalable, accurate and flexible predictive models, and inverse estimation approaches with applications in physics, materials science and engineering. He received the SIAM activity group on uncertainty quantification (SIAG/UQ) early career prize in 2022.