# A Whittle Index Approach to Minimizing Functions of Age of Information

arXiv: https://arxiv.org/abs/1908.10438 · HTML: https://mlsanity.com/paper/1908.10438/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Aug 2019.

## Built on by (most cited first, 21 of 21)

- [NeurWIN: Neural Whittle Index Network For Restless Bandits Via Deep RL](https://mlsanity.com/paper/2110.02128/) (2021)
- [Indexability is Not Enough for Whittle: Improved, Near-Optimal Algorithms for Restless Bandits](https://mlsanity.com/paper/2211.00112/) (2022)
- [An Online Learning Approach to Optimizing Time-Varying Costs of AoI](https://mlsanity.com/paper/2105.13383/) (2021)
- [Timely Communications for Remote Inference](https://mlsanity.com/paper/2404.16281/) (2024)
- [The Bandit Whisperer: Communication Learning for Restless Bandits](https://mlsanity.com/paper/2408.05686/) (2024)
- [Lagrangian Index Policy for Restless Bandits with Average Reward](https://mlsanity.com/paper/2412.12641/) (2024)
- [How Does Data Freshness Affect Real-time Supervised Learning?](https://mlsanity.com/paper/2208.06948/) (2022)
- [Deep Index Policy for Multi-Resource Restless Matching Bandit and Its Application in Multi-Channel Scheduling](https://mlsanity.com/paper/2408.07205/) (2024)
- [Tabular and Deep Reinforcement Learning for Gittins Index](https://mlsanity.com/paper/2405.01157/) (2024)
- [GINO-Q: Learning an Asymptotically Optimal Index Policy for Restless Multi-armed Bandits](https://mlsanity.com/paper/2408.09882/) (2024)
- [Online Learning of Whittle Indices for Restless Bandits with Non-Stationary Transition Kernels](https://mlsanity.com/paper/2506.18186/) (2025)
- [Multimodal Remote Inference](https://mlsanity.com/paper/2508.07555/) (2025)
- [Minimizing Age of Information for Mobile Edge Computing Systems: A Nested Index Approach](https://mlsanity.com/paper/2307.01366/) (2023)
- [Learning-based Scheduling for Information Accuracy and Freshness in Wireless Networks](https://mlsanity.com/paper/2310.15705/) (2023)
- [Age-of-information minimization under energy harvesting and non-stationary environment](https://mlsanity.com/paper/2504.04916/) (2025)
- [On the Regularity and Fairness of Combinatorial Multi-Armed Bandit](https://mlsanity.com/paper/2509.12457/) (2025)
- [Online Learning for Optimizing AoI-Energy Tradeoff under Unknown Channel Statistics](https://mlsanity.com/paper/2509.18654/) (2025)
- [Adaptive Requesting in Decentralized Edge Networks via Non-Stationary Bandits](https://mlsanity.com/paper/2601.08760/) (2026)
- [Real-Time Hard Peak Age-of-Information Safety with No-Regret Learning](https://mlsanity.com/paper/2607.27626/) (2026)
- [Threshold Structure of Optimal Policies in Restart POMDPs](https://mlsanity.com/paper/2608.10936/) (2026)
- [Learning When to Update: A Near-Optimal Timing Bandit Approach](https://mlsanity.com/paper/2609.37932/) (2026)

## Abstract (arXiv)

We consider a setting where multiple active sources send real-time updates over a single-hop wireless broadcast network to a monitoring station. Our goal is to design a scheduling policy that minimizes the time-average of general non-decreasing cost functions of Age of Information. We use a Whittle index based approach to find low complexity scheduling policies that have good performance. We prove that for a system with two sources, having possibly different cost functions and reliable channels, the Whittle index policy is exactly optimal. We derive structural properties of an optimal policy, that suggest that the performance of the Whittle index policy may be close to optimal in general. These results might also be of independent interest in the study of restless multi-armed bandit problems with similar underlying structure. We further establish that minimizing monitoring error for linear time-invariant systems and symmetric Markov chains is equivalent to minimizing appropriately chosen monotone functions of Age of Information. Finally, we provide simulations comparing the Whittle index policy with optimal scheduling policies found using dynamic programming, which support our results.