Back to the program

Contributed Papers

Bias Adjusted Calibration and Online Learning

Thu, Aug 6, 10:30 AM - 12:20 PM Room CC-103 Thomas M. Menino Convention & Exhibition Center

Xiaoxi ShenChair

Section on Statistical Learning and Data Science

6 Presentations

Advances in Budget-Split Experimentation for the Online Ads Marketplace

10:35 AM - 10:50 AM

Shan Ba (LinkedIn)

A/B Test Interference Bias BST

Bias-adjusted random forests for quantile estimation and prediction intervals

10:50 AM - 11:05 AM

Matt Berkowitz (Simon Fraser University (SFU))

Co-authors: Thomas Loughin (Simon Fraser University), Rachel Altman (Simon Fraser University)

random forest quantile regression forest bias correction quantile estimation prediction intervals

Convergence of Expected Suboptimality and Asymptotic Distribution of ADAM-Type Optimizers

11:05 AM - 11:20 AM

Co-authors: Ran Chen (Washington University in St. Louis)

Adam optimizer expected regret asymptotic normality stochastic approximation Polyak-Juditsky averaging

Error-Sensitive Dynamic Calibration for Online Optimization with Switching Costs

11:20 AM - 11:35 AM

Zsolt Talata (University of Kansas)

Co-authors: Toye Oloko (University of Kansas), Morteza Hashemi (University of Kansas)

Online convex optimization Error-sensitive Dynamic calibration

Likelihood-Free Adaptive Bayesian Inference via Nonparametric Distribution Matching

11:35 AM - 11:50 AM

Wenhui Sophia Lu (Stanford University)

Co-authors: Wing-Hung Wong (Stanford University)

simulator-based inference generative modeling approximate Bayesian computation nonparametric distribution matching Wasserstein distance likelihood-free inference

Reducing Sim-to-Real Gap in Bayesian Optimization with Online Simulator Calibration

11:50 AM - 12:05 PM

Piotr Suder (Duke University)

Co-authors: Eric Laber, Simon Mak (Duke University)

sim-to-real gap online calibration surrogate modeling deep Gaussian processes Bayesian optimization

↑