English

Analysis of Value Iteration Through Absolute Probability Sequences

Machine Learning 2025-02-06 v1

Abstract

Value Iteration is a widely used algorithm for solving Markov Decision Processes (MDPs). While previous studies have extensively analyzed its convergence properties, they primarily focus on convergence with respect to the infinity norm. In this work, we use absolute probability sequences to develop a new line of analysis and examine the algorithm's convergence in terms of the L2L^2 norm, offering a new perspective on its behavior and performance.

Keywords

Cite

@article{arxiv.2502.03244,
  title  = {Analysis of Value Iteration Through Absolute Probability Sequences},
  author = {Arsenii Mustafin and Sebastien Colla and Alex Olshevsky and Ioannis Ch. Paschalidis},
  journal= {arXiv preprint arXiv:2502.03244},
  year   = {2025}
}

Comments

8 pages

R2 v1 2026-06-28T21:33:33.651Z