Aging Bandits: Regret Analysis and Order-Optimal Learning Algorithm for Wireless Networks with Stochastic Arrivals

TL;DR

This paper studies a wireless network where channels' reliability is unknown and must be learned to minimize the Age-of-Information, introducing a new algorithm with bounded regret that outperforms traditional bandit solutions.

Contribution

It introduces a novel learning algorithm with bounded AoI regret, improving over existing methods that have logarithmic regret, for optimizing information freshness in wireless networks.

Findings

Existing bandit algorithms have logarithmic AoI regret.

The proposed algorithm achieves constant AoI regret.

First known algorithm with bounded AoI regret.

Abstract

We consider a single-hop wireless network with sources transmitting time-sensitive information to the destination over multiple unreliable channels. Packets from each source are generated according to a stochastic process with known statistics and the state of each wireless channel (ON/OFF) varies according to a stochastic process with unknown statistics. The reliability of the wireless channels is to be learned through observation. At every time slot, the learning algorithm selects a single pair (source, channel) and the selected source attempts to transmit its packet via the selected channel. The probability of a successful transmission to the destination depends on the reliability of the selected channel. The goal of the learning algorithm is to minimize the Age-of-Information (AoI) in the network over $T$ time slots. To analyze the performance of the learning algorithm, we introduce the notion of AoI regret, which is the difference between the expected cumulative AoI of the learning algorithm under consideration and the expected cumulative AoI of a genie algorithm that knows the reliability of the channels a priori. The AoI regret captures the penalty incurred by having to learn the statistics of the channels over the $T$ time slots. The results are two-fold: first, we consider learning algorithms that employ well-known solutions to the stochastic multi-armed bandit problem (such as $\epsilon$-Greedy, Upper Confidence Bound, and Thompson Sampling) and show that their AoI regret scales as $\Theta(\log T)$; second, we develop a novel learning algorithm and show that it has $O(1)$ regret. To the best of our knowledge, this is the first learning algorithm with bounded AoI regret.