1. the complete title of one (or more) paper(s) published in the open literature describing the work that the author claims describes a human-competitive result;

	Zhang, 


2. the name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s);
	Xiao-Cheng Liao,    xiaocheng@ecs.vuw.ac.nz    +64 27 888 3638
	School of Engineering and Computer Science, MARU 103, Victoria University of Wellington, Gate 7, Kelburn Parade, Wellington, New Zealand

	Yi Mei,   yi.mei@ecs.vuw.ac.nz   +64 4 886 5331  
	School of Engineering and Computer Science, Room CO358, Cotton Building, Victoria University of Wellington, Gate 7, Kelburn Parade, Wellington, New Zealand

	Mengjie Zhang,    Mengjie.Zhang@ecs.vuw.ac.nz   +64 4 4635 654
	School of Engineering and Computer Science, Room CO358, Cotton Building, Victoria University of Wellington, Gate 7, Kelburn Parade, Wellington, New Zealand


3. the name of the corresponding author (i.e., the author to whom notices will be sent concerning the competition);
	Yi Mei,   yi.mei@ecs.vuw.ac.nz   +64 4 886 5331 

4. the abstract of the paper(s);

	The control of traffic signals is crucial for improving transportation efficiency. Recently, learning-based methods, especially Deep Reinforcement Learning (DRL), garnered substantial success in the quest for more efficient traffic signal control strategies. However, the design of rewards in DRL highly demands domain knowledge to converge to an effective policy, and the final policy also presents difficulties in terms of explainability. In this work, a new learning-based method for signal control in complex intersections is proposed. In our approach, we design a concept of phase urgency for each signal phase. During signal transitions, the traffic light control strategy selects the next phase to be activated based on the phase urgency. We then proposed to represent the urgency function as an explainable tree structure. The urgency function can calculate the phase urgency for a specific phase based on the current road conditions. Genetic programming is adopted to perform gradient-free optimization of the urgency function. We test our algorithm on multiple public traffic signal control datasets. The experimental results indicate that the tree-shaped urgency function evolved by genetic programming outperforms the baselines, including a state-of-the-art method in the transportation field and a well-known DRL-based method.

5. a list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria (see above) that the author claims that the work satisfies;

	B, C, D, E, G

6. A statement stating why the result satisfies the criteria that the contestant 
claims (see examples of statements of human-competitiveness as a guide to aid in 
constructing this part of the submission).

re: (B) "The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal."

	Our algorithm outperform the state-of-the-art heuristic algorithm (published in Transportation Research Part C in 2013) in the transportation field as well as MPLight (published on AAAI 2000), a DRL-based algorithm and state-of-the-art learning-based method for large-scale traffic networks.

re: (C) "The result is equal to or better than a result that was placed into a database or archive of results maintained by an internationally recognized panel of scientific experts."

	Our dataset comes from a publicly available real-world traffic dataset [1] and our method outperforms the results maintained by a research team at Pennsylvania State University.

	[1] https://traffic-signal-control.github.io/

re: (D): "The result is publishable in its own right as a new scientific result independent of the fact that the result was mechanically created."

	We propose a method to utilize genetic programming to address an important real world problem, traffic signal control with complete phase in large-scale road networks. Our method was validated using the famous traffic flow simulation engine, CityFlow [2], with several traffic datasets collected from real-world traffic scenarios. Our method reduces the average travel time of vehicles by up to 13.52%. This improvement can lead to CO2 emissions and traffic congestion reduction, economic savings, commute time savings that improve drivers’ mental health, etc. Besides, from a technical perspective, our method does not require complex lane features and the final traffic signal control strategy is explainable. Undoubtedly, our technique holds significant promise for practical application. When we conduct real-world trials, the results will be publishable.

	The results of this paper are promising and the paper has been accepted as a full paper by GECCO 2024 and been nominated for the GP Track Best Paper Award [3].
	This paper achieved an average reviewer recommendation score of 4.2, with some of their feedback highlights as follows:
	- "Generally, the proposed method is interesting and has significant research value, and the presentation is clear"
	- "I thought this paper was great overall. The task is well-laid out and the motivation and prior work are well-established and made clear. Overall it was a pleasure to read."
	- "A very interesting study and read on a nice problem. The experimental design appears strong and includes benchmarking performance against a number of other approaches."
	- "Overall, the authors claim that their approach improves efficiency and explainability, which is essential in real-world problems."

	[2] https://cityflow-project.github.io/
	[3] https://gecco-2024.sigevo.org/Best-Paper-Nominations


re: (E): "The result is equal to or better than the most recent human-created solution to a long-standing problem for which there has been a succession of increasingly better human-created solutions."

	Our proposed GPLight significantly outperforms the state-of-the-art algorithm in the transportation field, Max-Pressure, in terms of average vehicle travel time (relatively reduces vehicle travel time by 13.52% and see this in the paper), on multiple real-world traffic scenario datasets. Max-Pressure is a human-created, theoretically supported, and explainable heuristic algorithm to dynamically control traffic lights.

re: (G): "The result solves a problem of indisputable difficulty in its field."

	Despite the significant advancements achieved by Deep Reinforcement Learning (DRL) methods in traffic signal control, their dependence on neural networks brings about new difficulties. These difficulties include a lack of explainability and incompatibility with edge devices. Consequently, in the real world, most traffic signal control methods still depend on traditional approaches. Even in developed countries like the United States, the proportion of intelligent traffic signals remains below 5% [4]. GPLight is able to find explainable rule-based TSC strategies that features a human-understandable and low-power deployment model. Our proposed method has significant advantages over DRL methods in terms of explainability and compatibility with edge devices, making it more feasible than DRL-based methods for application in real-world traffic scenarios.

	[4] Tang, Keshuang, Manfred Boltze, Hideki Nakamura, and Zong Tian. Global practices on road traffic signal control: Fixed-time control at isolated intersections. Elsevier, 2019.

7. a full citation of the paper (that is, author names; title, publication date; name of journal, conference, or book in which article appeared; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable);

	Xiao-Cheng Liao, Yi Mei, and Mengjie Zhang. 2024. Learning Traffic Signal Control via Genetic Programming. In Genetic and Evolutionary Computation Conference (GECCO ’24), July 14–18, 2024, Melbourne, VIC, Australia. ACM, New York, NY, USA, 9 pages. https://doi.org/10.1145/3638529.3654037

8. a statement either that "any prize money, if any, is to be divided equally among the co-authors" OR a specific percentage breakdown as to how the prize money, if any, is to be divided among the co-authors;

	All prize money, if any, will be allocated to the primary author to support his PhD studies.

9. a statement stating why the authors expect that their entry would be the "best".

	Firstly, this paper aims to solve an extremely important real-world problem, traffic signal control. Traffic signal control plays a crucial role in coordinating traffic flows with different directions and reducing congestion.  Our proposed approach further reduces the average vehicle travel time by 13.52% compared to the human-designed method. This can be translated to reduce prolonged waiting and frequent stops, leading to lower fuel consumption and CO2 emissions, which significantly benefit the environment by alleviating air pollution. More importantly, optimized traffic signal control can boost societal productivity and reduce the costs associated with traffic management and accident response, resulting in significant socio-economic benefits.

	Secondly, this paper proposes a method to evolve human-understandable strategies for complete 8-phase multi-intersection traffic signal control. Our method does not require the design of intricate reward functions like DRL does which involves a significant amount of domain knowledge and expertise. This means that practitioners do not need extra experience and expertise from other field to use our algorithm, which can greatly facilitate the practical application of the research.

	Lastly, most existing learning-based methods rely on complex deep models, which limits their interpretability and deployment on edge devices. However, explainable traffic signal control strategies are crucial. The absence of transparency for signal control strategies poses significant barriers to establishing trust from users and for dispatchers to examine potential weaknesses in the policies. The final traffic signal control strategy of our method is rule-based, characterized by interpretability and low-cost deployment. This can also greatly facilitate its implementation in real-world traffic scenarios.

10. An indication of the general type of genetic or evolutionary computation used, such as GA (genetic algorithms), GP (genetic programming), ES (evolution strategies), EP (evolutionary programming), LCS (learning classifier systems), GI (genetic improvement), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc.

	GP (genetic programming)

11. The date of publication of each paper. If the date of publication is not on or before the deadline for submission, but instead, the paper has been unconditionally accepted for publication and is “in press” by the deadline for this competition, the entry must include a copy of the documentation establishing that the paper meets the "in press" requirement.

In-press for GECCO 2024. See below accepatance email: 

Dear Xiao-Cheng Liao,

Congratulations! Your paper
(pap254s2) Learning Traffic Signal Control via Genetic Programming

has been accepted as a *full paper* (page limit: 8 pages excluding references) for GECCO 2024.

Reviews are now available in the submission website at https://ssl.linklings.net/conferences/gecco .

Please confirm or decline acceptance there. Acceptance is subject to the condition that you consider the comments of the reviewers when preparing the camera-ready version of your manuscript.

After confirming acceptance, you will receive a separate email with instructions and information about the copyright process. You need to complete the copyright form to get the copyright notice with the final DOI you have to include in the camera-ready version of the manuscript. Your camera-ready version must be submitted via the submission system (https://ssl.linklings.net/conferences/gecco/) by Thursday, April 11th, 2024.

Please, prepare your camera-ready version following the instructions and templates at https://gecco-2024.sigevo.org/Paper-Submission-Instructions (see the Camera-Ready Instructions). Upload preliminary versions of your manuscript prior to the deadline to check if there is any problem or formatting issue (e.g., inappropriate or not embedded fonts).

Please note that May 10th, 2024 is the firm deadline for authors of accepted papers/posters to register. At least one author for each accepted paper/poster must be registered by then, and at least one author must present the paper at the conference. The registration site will open within the next two weeks at https://gecco-2024.sigevo.org/Registration

Thank you for submitting your work to GECCO 2024 and we look forward to seeing you at the conference!

Best wishes,
Julia Handl
Editor-in-Chief