TY - GEN
T1 - Conditioning Sparse Variational Gaussian Processes for Online Decision-making
AU - Maddox, Wesley J.
AU - Stanton, Samuel
AU - Wilson, Andrew Gordon
N1 - Funding Information:
WJM, SS, AGW are supported by an Amazon Research Award, NSF I-DISRE 193471, NIH R01 DA048764-01A1, NSF IIS-1910266, and NSF 1922658 NRT-HDR: FUTURE Foundations, Translation, and Responsibility for Data Science. WJM was additionally supported by an NSF Graduate Research Fellowship under Grant No. DGE-1839302. SS is additionally supported by the United States Department of Defense through the National Defense Science & Engineering Graduate (ND-SEG) Fellowship Program. We'd like to thank Greg Benton for setting up the volatility experiment and for helpful discussions and Nate Gruver and Eytan Bakshy for helpful comments.
Publisher Copyright:
© 2021 Neural information processing systems foundation. All rights reserved.
PY - 2021
Y1 - 2021
N2 - With a principled representation of uncertainty and closed form posterior updates, Gaussian processes (GPs) are a natural choice for online decision making. However, Gaussian processes typically require at least O(n2) computations for n training points, limiting their general applicability. Stochastic variational Gaussian processes (SVGPs) can provide scalable inference for a dataset of fixed size, but are difficult to efficiently condition on new data. We propose online variational conditioning (OVC), a procedure for efficiently conditioning SVGPs in an online setting that does not require re-training through the evidence lower bound with the addition of new data. OVC enables the pairing of SVGPs with advanced look-ahead acquisition functions for black-box optimization, even with non-Gaussian likelihoods. We show OVC provides compelling performance in a range of applications including active learning of malaria incidence, and reinforcement learning on MuJoCo simulated robotic control tasks.
AB - With a principled representation of uncertainty and closed form posterior updates, Gaussian processes (GPs) are a natural choice for online decision making. However, Gaussian processes typically require at least O(n2) computations for n training points, limiting their general applicability. Stochastic variational Gaussian processes (SVGPs) can provide scalable inference for a dataset of fixed size, but are difficult to efficiently condition on new data. We propose online variational conditioning (OVC), a procedure for efficiently conditioning SVGPs in an online setting that does not require re-training through the evidence lower bound with the addition of new data. OVC enables the pairing of SVGPs with advanced look-ahead acquisition functions for black-box optimization, even with non-Gaussian likelihoods. We show OVC provides compelling performance in a range of applications including active learning of malaria incidence, and reinforcement learning on MuJoCo simulated robotic control tasks.
UR - http://www.scopus.com/inward/record.url?scp=85128049825&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85128049825&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85128049825
T3 - Advances in Neural Information Processing Systems
SP - 6365
EP - 6379
BT - Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
A2 - Ranzato, Marc'Aurelio
A2 - Beygelzimer, Alina
A2 - Dauphin, Yann
A2 - Liang, Percy S.
A2 - Wortman Vaughan, Jenn
PB - Neural information processing systems foundation
T2 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
Y2 - 6 December 2021 through 14 December 2021
ER -