Probabilistic Surrogate Networks for Simulators with Unbounded Randomness
Probabilistic Surrogate Networks for Simulators with Unbounded Randomness
About this item
Full title
Author / Creator
Publisher
Ithaca: Cornell University Library, arXiv.org
Journal title
Language
English
Formats
Publication information
Publisher
Ithaca: Cornell University Library, arXiv.org
Subjects
More information
Scope and Contents
Contents
We present a framework for automatically structuring and training fast, approximate, deep neural surrogates of stochastic simulators. Unlike traditional approaches to surrogate modeling, our surrogates retain the interpretable structure and control flow of the reference simulator. Our surrogates target stochastic simulators where the number of rand...
Alternative Titles
Full title
Probabilistic Surrogate Networks for Simulators with Unbounded Randomness
Authors, Artists and Contributors
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2310134126
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2310134126
Other Identifiers
E-ISSN
2331-8422
