Saved in:
Bibliographic Details
Main Authors: Morton, David P., Dowson, Oscar, Pagnoncelli, Bernardo K.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.22981
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914454541697024
author Morton, David P.
Dowson, Oscar
Pagnoncelli, Bernardo K.
author_facet Morton, David P.
Dowson, Oscar
Pagnoncelli, Bernardo K.
contents We study a class of multi-stage stochastic programs, which incorporate modeling features from Markov decision processes (MDPs). This class includes structured MDPs with continuous action and state spaces. We extend policy graphs to include decision-dependent uncertainty for one-step transition probabilities as well as a limited form of statistical learning. We focus on the expressiveness of our modeling approach, illustrating ideas with a series of examples of increasing complexity. As a solution method, we develop new variants of stochastic dual dynamic programming, including approximations to handle non-convexities.
format Preprint
id arxiv_https___arxiv_org_abs_2509_22981
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MDP modeling for multi-stage stochastic programs
Morton, David P.
Dowson, Oscar
Pagnoncelli, Bernardo K.
Machine Learning
Optimization and Control
90C15, 90C40
We study a class of multi-stage stochastic programs, which incorporate modeling features from Markov decision processes (MDPs). This class includes structured MDPs with continuous action and state spaces. We extend policy graphs to include decision-dependent uncertainty for one-step transition probabilities as well as a limited form of statistical learning. We focus on the expressiveness of our modeling approach, illustrating ideas with a series of examples of increasing complexity. As a solution method, we develop new variants of stochastic dual dynamic programming, including approximations to handle non-convexities.
title MDP modeling for multi-stage stochastic programs
topic Machine Learning
Optimization and Control
90C15, 90C40
url https://arxiv.org/abs/2509.22981