Bayesian Stochastic Gradient Descent for Stochastic Optimization with Streaming Input Data
From MaRDI portal
Abstract: We consider stochastic optimization under distributional uncertainty, where the unknown distributional parameter is estimated from streaming data that arrive sequentially over time. Moreover, data may depend on the decision of the time when they are generated. For both decision-independent and decision-dependent uncertainties, we propose an approach to jointly estimate the distributional parameter via Bayesian posterior distribution and update the decision by applying stochastic gradient descent on the Bayesian average of the objective function. Our approach converges asymptotically over time and achieves the convergence rates of classical SGD in the decision-independent case. We demonstrate the empirical performance of our approach on both synthetic test problems and a classical newsvendor problem.
Recommendations
- Distribution-free algorithms for predictive stochastic programming in the presence of streaming data
- Bayesian Distributionally Robust Optimization
- A Bayesian approach to data-driven multi-stage stochastic optimization
- A Bayesian risk approach to data-driven stochastic optimization: formulations and asymptotics
- Stochastic optimization on Bayesian nets
Cites work
- ``Stack them high, let 'em fly: lot-sizing policies when inventories stimulate demand
- A Bayesian risk approach to data-driven stochastic optimization: formulations and asymptotics
- A class of stochastic programs with decision dependent uncertainty
- A stochastic inventory model with stock dependent demand items.
- About the non-asymptotic behaviour of Bayes estimators
- Adaptive subgradient methods for online learning and stochastic optimization
- An approximate dynamic programming framework for modeling global climate policy under decision-dependent uncertainty
- An inventory model for deteriorating items with stock-dependent demand rate
- Analysis of Langevin Monte Carlo via convex optimization
- Augmented nested sampling for stochastic programs with recourse and endogenous uncertainty
- Bayesian Distributionally Robust Optimization
- Convergence rates of posterior distributions for non iid observations
- Data-driven chance constrained stochastic program
- Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
- Data-driven robust optimization
- Decision-dependent probabilities in stochastic programs with recourse
- Distributionally Robust Control of Constrained Stochastic Systems
- Distributionally Robust Convex Optimization
- Distributionally robust optimization under moment uncertainty with application to data-driven problems
- Distributionally robust optimization with decision dependent ambiguity sets
- Fast Langevin based algorithm for MCMC in high dimensions
- High-dimensional Bayesian inference via the unadjusted Langevin algorithm
- scientific article; zbMATH DE number 42073 (Why is no real title available?)
- scientific article; zbMATH DE number 1220667 (Why is no real title available?)
- scientific article; zbMATH DE number 1972910 (Why is no real title available?)
- scientific article; zbMATH DE number 1569102 (Why is no real title available?)
- Lectures on stochastic programming. Modeling and theory
- Multistage distributionally robust mixed-integer programming with decision-dependent moment-based ambiguity sets
- Newsvendor-type models with decision-dependent uncertainty
- Numerical integration of the Langevin equation: Monte Carlo simulation
- On some recent advances on high dimensional Bayesian statistics
- Online learning and online convex optimization
- Optimization under decision-dependent uncertainty
- Penalty Functions and Duality in Stochastic Programming Via ϕ-Divergence Functionals
- Probability
- Quantitative Stability in Stochastic Programming: The Method of Probability Metrics
- Robust ranking and selection with optimal computing budget allocation
- Simulation Budget Allocation for Selecting the Top-m Designs With Input Uncertainty
- Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic Programming
- Wasserstein Distributionally Robust Stochastic Control: A Data-Driven Approach
- What you should know about simulation and derivatives
Cited in
(8)- Distribution-free algorithms for predictive stochastic programming in the presence of streaming data
- Stochastic Optimization with Decision-Dependent Distributions
- A Bayesian approach to data-driven multi-stage stochastic optimization
- Stochastic approximation for multi-period simulation optimization with streaming input data
- Supercharging digital twins with AI
- Blackbox simulation optimization
- Infinitesimal perturbation analysis (IPA) derivative estimation with unknown parameters
- Simulation-based dynamic optimization for ambulance dispatch and casualty distribution in mass casualty incident
This page was built for publication: Bayesian Stochastic Gradient Descent for Stochastic Optimization with Streaming Input Data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6188508)
