Randomized Multilevel Monte Carlo for Embarrassingly Parallel Inference
Name:
2107.01913v2.pdf
Size:
1.027Mb
Format:
PDF
Description:
Accepted Manuscript
Embargo End Date:
2023-03-10
Type
Conference PaperKAUST Department
Computer, Electrical and Mathematical Sciences and Engineering Division, King Abdullah University of Science and Technology, Thuwal, 23955, Kingdom of Saudi ArabiaComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
Statistics
Date
2022-03-10Embargo End Date
2023-03-10Permanent link to this record
http://hdl.handle.net/10754/670196
Metadata
Show full item recordAbstract
This position paper summarizes a recently developed research program focused on inference in the context of data centric science and engineering applications, and forecasts its trajectory forward over the next decade. Often one endeavours in this context to learn complex systems in order to make more informed predictions and high stakes decisions under uncertainty. Some key challenges which must be met in this context are robustness, generalizability, and interpretability. The Bayesian framework addresses these three challenges, while bringing with it a fourth, undesirable feature: it is typically far more expensive than its deterministic counterparts. In the 21st century, and increasingly over the past decade, a growing number of methods have emerged which allow one to leverage cheap low-fidelity models in order to precondition algorithms for performing inference with more expensive models and make Bayesian inference tractable in the context of high-dimensional and expensive models. Notable examples are multilevel Monte Carlo (MLMC), multi-index Monte Carlo (MIMC), and their randomized counterparts (rMLMC), which are able to provably achieve a dimension-independent (including ∞- dimension) canonical complexity rate with respect to mean squared error (MSE) of 1/MSE. Some parallelizability is typically lost in an inference context, but recently this has been largely recovered via novel double randomization approaches. Such an approach delivers independent and identically distributed samples of quantities of interest which are unbiased with respect to the infinite resolution target distribution. Over the coming decade, this family of algorithms has the potential to transform data centric science and engineering, as well as classical machine learning applications such as deep learning, by scaling up and scaling out fully Bayesian inference.Citation
Jasra, A., Law, K. J. H., Tarakanov, A., & Yu, F. (2022). Randomized Multilevel Monte Carlo for Embarrassingly Parallel Inference. Communications in Computer and Information Science, 3–21. https://doi.org/10.1007/978-3-030-96498-6_1Sponsors
KJHL and AT were supported by The Alan Turing Institute under the EPSRC grant EP/N510129/1. AJ and FY acknowledge KAUST baseline support.Publisher
Springer International PublishingConference/Event name
21st Smoky Mountains Computational Sciences and Engineering Conference, SMC 2021ISBN
97830309649799783030964986
arXiv
2107.01913Additional Links
https://link.springer.com/10.1007/978-3-030-96498-6_1ae974a485f413a2113503eed53cd6c53
10.1007/978-3-030-96498-6_1