English

Zero-one Laws for a Control Problem with Random Action Sets

Optimization and Control 2024-04-11 v1 Probability

Abstract

In many control problems there is only limited information about the actions that will be available at future stages. We introduce a framework where the Controller chooses actions a0,a1,a_{0}, a_{1}, \ldots, one at a time. Her goal is to maximize the probability that the infinite sequence (a0,a1,)(a_{0}, a_{1}, \ldots) is an element of a given subset GG of NN\mathbb{N}^{\mathbb{N}}. The set GG, called the goal, is assumed to be a Borel tail set. The Controller's choices are restricted: having taken a sequence ht=(a0,,at1)h_{t} = (a_{0}, \ldots, a_{t-1}) of actions prior to stage tNt \in \mathbb{N}, she must choose an action ata_{t} at stage tt from a non-empty, finite subset A(ht)A(h_{t}) of N\mathbb{N}. The set A(ht)A(h_{t}) is chosen from a distribution ptp_{t}, independently over all tNt \in \mathbb{N} and all htNth_{t} \in \mathbb{N}^{t}. We consider several information structures defined by how far ahead into the future the Controller knows what actions will be available. In the special case where all the action sets are singletons (and thus the Controller is a dummy), Kolmogorov's 0-1 law says that the probability for the goal to be reached is 0 or 1. We construct a number of counterexamples to show that in general the value of the control problem can be strictly between 0 and 1, and derive several sufficient conditions for the 0-1 ``law" to hold.

Cite

@article{arxiv.2404.07012,
  title  = {Zero-one Laws for a Control Problem with Random Action Sets},
  author = {János Flesch and Arkadi Predtetchinski and William D Sudderth and Xavier Venel},
  journal= {arXiv preprint arXiv:2404.07012},
  year   = {2024}
}
R2 v1 2026-06-28T15:49:57.681Z