# Quantifying the Impact of Non-Stationarity in Reinforcement Learning-Based Traffic Signal Control

arXiv: https://arxiv.org/abs/2004.04778 · HTML: https://mlsanity.com/paper/2004.04778/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Apr 2020.

## Built on by (most cited first, 6 of 6)

- [The Max-Min Formulation of Multi-Objective Reinforcement Learning: From Theory to a Model-Free Algorithm](https://mlsanity.com/paper/2406.07826/) (2024)
- [The Real Deal: A Review of Challenges and Opportunities in Moving Reinforcement Learning-Based Traffic Signal Control Systems Towards Reality](https://mlsanity.com/paper/2206.11996/) (2022)
- [Improving Intrinsic Exploration by Creating Stationary Objectives](https://mlsanity.com/paper/2310.18144/) (2023)
- [Towards Real-World Deployment of Reinforcement Learning for Traffic Signal Control](https://mlsanity.com/paper/2103.16223/) (2021)
- [Deep Reinforcement Learning-based Intelligent Traffic Signal Controls with Optimized CO2 emissions](https://mlsanity.com/paper/2310.13129/) (2023)
- [Counterfactual experience augmented off-policy reinforcement learning](https://mlsanity.com/paper/2503.13842/) (2025)

## Abstract (arXiv)

In reinforcement learning (RL), dealing with non-stationarity is a challenging issue. However, some domains such as traffic optimization are inherently non-stationary. Causes for and effects of this are manifold. In particular, when dealing with traffic signal controls, addressing non-stationarity is key since traffic conditions change over time and as a function of traffic control decisions taken in other parts of a network. In this paper we analyze the effects that different sources of non-stationarity have in a network of traffic signals, in which each signal is modeled as a learning agent. More precisely, we study both the effects of changing the textitcontext in which an agent learns (e.g., a change in flow rates experienced by it), as well as the effects of reducing agent observability of the true environment state. Partial observability may cause distinct states (in which distinct actions are optimal) to be seen as the same by the traffic signal agents. This, in turn, may lead to sub-optimal performance. We show that the lack of suitable sensors to provide a representative observation of the real state seems to affect the performance more drastically than the changes to the underlying traffic patterns.