Achieving Efficiency in Black-Box Simulation of Distribution Tails with Self-Structuring Importance Samplers

• DOI: 10.1287/opre.2021.0331
• Publication Date: 2023-07-19T13:51:30Z
• Authors: Anand Deo; Karthyek Murthy
• Keywords: Stylized fact; Black box; Structuring; Variance reduction

Scalable and efficient importance sampling for managing tail risks As the models employed in realm of risk analytics optimization become increasingly sophisticated, it is crucial that management tools, such as variance reduction techniques, are typically designed stylized on a case by basis evolve to scale well gain broader applicability. In paper titled “Achieving efficiency black-box simulation distribution tails with self-structuring samplers,” authors take step toward this goal introducing novel scheme estimating objectives modeled diverse range including linear programs, integer feature maps specified neural networks, etc. Instead explicitly tailoring change each specific model, conventionally done, identifies an elementary transformation samples. This transformation, when applied alike across wide variety models, yields near-optimal estimating/optimizing over expectations.