Saved in:
Bibliographic Details
Main Authors: Borrell, Enric Ribera, Richter, Lorenz, Schütte, Christof
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.00962
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916897545519104
author Borrell, Enric Ribera
Richter, Lorenz
Schütte, Christof
author_facet Borrell, Enric Ribera
Richter, Lorenz
Schütte, Christof
contents We extend the standard reinforcement learning framework to random time horizons. While the classical setting typically assumes finite and deterministic or infinite runtimes of trajectories, we argue that multiple real-world applications naturally exhibit random (potentially trajectory-dependent) stopping times. Since those stopping times typically depend on the policy, their randomness has an effect on policy gradient formulas, which we (mostly for the first time) derive rigorously in this work both for stochastic and deterministic policies. We present two complementary perspectives, trajectory or state-space based, and establish connections to optimal control theory. Our numerical experiments demonstrate that using the proposed formulas can significantly improve optimization convergence compared to traditional approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00962
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Reinforcement Learning with Random Time Horizons
Borrell, Enric Ribera
Richter, Lorenz
Schütte, Christof
Machine Learning
Optimization and Control
Probability
We extend the standard reinforcement learning framework to random time horizons. While the classical setting typically assumes finite and deterministic or infinite runtimes of trajectories, we argue that multiple real-world applications naturally exhibit random (potentially trajectory-dependent) stopping times. Since those stopping times typically depend on the policy, their randomness has an effect on policy gradient formulas, which we (mostly for the first time) derive rigorously in this work both for stochastic and deterministic policies. We present two complementary perspectives, trajectory or state-space based, and establish connections to optimal control theory. Our numerical experiments demonstrate that using the proposed formulas can significantly improve optimization convergence compared to traditional approaches.
title Reinforcement Learning with Random Time Horizons
topic Machine Learning
Optimization and Control
Probability
url https://arxiv.org/abs/2506.00962