Increasing Information for Model Predictive Control with Semi-Markov Decision Processes

Rémy Hosseinkhan Boucher, Onofrio Semeraro, Lionel Mathelin
1/28/2025
cs.LG

No official code found. Best implementation: ReHoss/lbmpc_semimarkov (2 stars, MIT, last commit 2024-05-18).

Abstract

Recent works in Learning-Based Model Predictive Control of dynamical systems show impressive sample complexity performances using criteria from Information Theory to accelerate the learning procedure. However, the sequential exploration opportunities are limited by the system local state, restraining the amount of information of the observations from the current exploration trajectory. This article resolves this limitation by introducing temporal abstraction through the framework of Semi-Markov Decision Processes. The framework increases the total information of the gathered data for a fixed sampling budget, thus reducing the sample complexity.

View on arXivView PDF

Code Implementations(1)

ReHoss/lbmpc_semimarkovCommunity100%
MIT

Repository for the article: "Increasing Information for Model Predictive Control with Semi-Markov Decision Processes"

21sed, ShellMay 11, 20242 years agoMIT

Similar Papers

An Optimal Policy for Learning Controllable Dynamics by Exploration

82% match2025

Learning a Fast Mixing Exogenous Block MDP using a Single Trajectory

80% match1 repo2024

Learning in Markov Decision Processes with Exogenous Dynamics

80% match2026

Time-Series-Informed Closed-loop Learning for Sequential Decision Making and Control

80% match2024

Time-Aware World Model for Adaptive Prediction and Control

79% match2 repos152026

Diffusion-Based Forecasting for Uncertainty-Aware Model Predictive Control

79% match2025

Cite this paper

@article{boucher2025increasing,
  title  = {Increasing Information for Model Predictive Control with Semi-Markov Decision Processes},
  author = {Rémy Hosseinkhan Boucher and Onofrio Semeraro and Lionel Mathelin},
  year   = {2025},
  eprint = {2501.17256},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2501.17256v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2501.17256.svg)](https://codeofpaper.com/paper/2501.17256)

Discussion