Regression-based Monte Carlo Integration
Event Type
Technical Paper
Interest Areas
Research & Education
Presentation Types
Registration Categories
Full Conference Supporter
Full Conference
Virtual Conference Supporter
Virtual Conference
Exhibitor Additional Full Conference
Exhibitor Full Conference
DescriptionMonte Carlo estimation involves computing the expected value (i.e., a constant function) from the stochastic evaluations of the integrand. We show that it is possible to use a more complex function than a constant one to construct a provably more efficient control-variate estimator for Monte Carlo integration.