Volume 16, Issue 3

newsletter of the ACM Special Interest Group on Genetic and Evolutionary Computation


Welcome to the 3rd (Fall) 2023 issue of the SIGEvolution newsletter! We start with an overview of the 2023 Humies competition, which awards results produced by evolutionary algorithms that are competitive or outperform results produced by human designers. Our next contribution sumarizes the winners of the 2022 SIGEVO Best Dissertation Award, which recognizes excellent research by doctoral candidates in the field of evolutionary computing. We complete the issue with recent announcements and forthcoming calls for papers. Please get in touch if you would like to contribute an article for a future issue or have suggestions for the newsletter.
Gabriela Ochoa, Editor

About the Cover

The cover image illustrates the work of the 2023 Humies Competition winners. The image combines Figures 5 – 7 from the CryptOpt article by Joel Kuepper et. al., where Genetic Improvement is used to generate fast implementations for cryptographic arithmetic. The images illustrate the algorithm implementing an example expression. The first step is to decompose it into simpler operations and connect them (left graph). The second step is to generate a valid order to implement those operations, as shown in the middle. It also shows the intervals, in which operations can be placed. For example, the light blue Add4 can be placed in its own position or the one before. Local search then randomly mutates the current representation, as seen on the right. In this example, it changes the position of Add4 and the instruction for Add1. This results in a different implementation that calculates the same expression. Both implementations are run on the hardware and the faster one is kept.

The 2023 Humies Awards

Erik Goodman, BEACON Centre, Michigan State University, USA

The GECCO 2023 Conference was held in hybrid mode again this year. It was attended by more than 600 people on site in Lisbon, Portugal, plus more than 200 attending virtually. It was held July 15-19, with the finalists in the Humies competition presenting in a plenary session on Tuesday, July 18 that was attended by more than 150 people. Eight finalists presented their work in 10-minute talks, or, in one case, a pre-recorded video

The Humies competition annually awards $10,000 in cash prizes for computational results that are deemed to be competitive with results produced by human beings, but are generated automatically by computers. This year, 19 entries were submitted, and the judges selected eight of those to be finalists, based on the papers and entry forms they submitted.

The judges viewed all presentations and then held their deliberations to select the prize winners. All 19 entries and the presentation materials of the eight finalists are available to the general public at the Humies website www.human-competitive.org.

The competition, sponsored annually by John Koza (who is widely regarded as the “Father of Genetic Programming”) solicits papers published within the last year that describe work fulfilling one or more of eight criteria, including such features as winning a regulated competition against humans or other programs; producing results that are publishable in their own right, not because they were created by a computer program; patentability; and other criteria described in full on the Humies website. This year, the Gold Award included US$5,000; the two Silver Awards, $2,000 each, and the Bronze, $1,000. The judges for the competition were Wolfgang Banzhaf, Erik Goodman, Una-May O’Reilly, Lee Spector and Darrell Whitley. Publicity for the Humies was done by Bill Langdon.

As usual, after considering all the entries, the judges wanted to give awards to many of them, but the final result was the following four awards:

Gold Award

The Gold, a $5,000 prize, was awarded to the team of Joel Kuepper, Andres Erbsen, Jason Gross, Owen Conoly, Chuyue Sun, Samuel Tian, David Wu, Adam Chlipala, Chitchanok Chuengsatiansun, Daniel Genkin, Markus Wagner and Yuval Yarom. Kuepper, who presented, is from the University of Adelaide (Australia), and his co-authors are from institutions including MIT, Stanford, Melbourne, Georgia Tech, Monash and Ruhr University Bochum.

Their paper was entitled “CryptOpt: Verified compilation with randomized program search for cryptographic primitives” and has been accepted for publication in PLDI, 2023.

Their CryptOpt represents the first compilation pipeline that specializes high-level cryptographic functional programs into assembly code that runs significantly faster than what GCC or Clang produce and is formally verified through the Fiat Cryptography framework with a newly extended program equivalence checker. The overall prototype is quite practical—e.g., it produces new fastest-known implementations of finite-field arithmetic for both Curve25519 (part of the TLS standard) and the Bitcoin elliptic curve secp256k1 for the Intel 12th and 13th generations.

Silver Award (Tie)

One of the Silver Awards and $2,000 went to a team which included Jordan MacLachlan (who presented), Yi Mei, Fangfang Zhang, Mengjie Zhang and Jessica Signal. The team comes from the University of Wellington (NZ) and from a Wellington organization that provides ambulance services.

Their paper is entitled, “Learning Emergency Medical Dispatch Policies via Genetic Programming,” and appears in the GECCO 2023 Proceedings.

The team proposed a modularised Genetic Programming Hyper Heuristic framework to learn the five key decisions of Emergency Medical Dispatch (EMD) within a reactive decision-making process.

They minimized the representational distance between their work and reality by working with their local ambulance service to design a set of heuristics approximating their current decision-making processes and a set of synthetic datasets influenced by existing patterns in practice. Through their modularized framework, they learned each decision independently to identify those most valuable to EMD and learned all five decisions simultaneously, improving performance by 69% on the largest novel dataset. They analyzed the decision-making logic behind several learned rules to further improve their understanding of EMD.

Silver Award (Tie)

The other Silver Award and $2,000 went to William G. La Cava, Paul C. Lee, Imran Ajmal, Xiruo Ding, Priyanka Solanki, Jordana B. Cohen, Jason H. Moore and Daniel S. Herman, a team from Boston Children’s Hospital, Harvard Medical School, and the University of Pennsylvania.

Bill La Cava presented their entry entitled “Symbolic Regression for Interpretable Clinical Prediction Models.” They adapted a symbolic regression method, coined the feature engineering automation tool (FEAT), to train concise and accurate, highly interpretable models from high-dimensional electronic health record (EHR) data.

They presented an in-depth application of FEAT to classify hypertension, hypertension with unexplained hypokalemia, and apparent treatment-resistant hypertension (aTRH) using EHR data for 1200 subjects receiving longitudinal care in a large healthcare system. FEAT models trained to predict phenotypes adjudicated by chart review had equivalent or higher discriminative performance (p < 0.001) and were at least three times smaller (p < 1 × 10^−6) than other potentially interpretable models. For aTRH, FEAT generated a six-feature, highly discriminative (positive predictive value = 0.70, sensitivity = 0.62), and clinically intuitive model.

Bronze Award

The Bronze Award and $1,000 went to the team of Julia Reuter, Hani Elmestikawy, Sanaz Mostaghim, Berend van Wachem, Fabien Evrard and Manoj Cendrollu, of Otto-von-Guericke-University, Magdeburg.

Julia Reuter presented their entry entitled, “Towards Improving Simulations of Flows around Spherical Particles Using Genetic Programming.” A longstanding challenge in the simulation of fluids is to include the flows around those containing particles, which to date has been restricted to including relatively few particles.

The team tackled the modeling of the forces on an individual particle in the presence, first of a single other particle, then in the presence of many. They used genetic programming to do symbolic regression of expressions to represent those forces and flows, but still under some simplifying assumptions. Progress in this field is an essential step in allowing the full simulation of flows of fluids such as blood.

ACM SIGEVO Best Dissertation Award 2023

Manuel López-Ibáñez, University of Manchester, UK

The SIGEVO Best Dissertation Award was created in 2019 to recognize excellent thesis research by doctoral candidates in the field of evolutionary computing. Doctoral dissertation awards are given by other Special Interest Groups of ACM, such as SIGCOMM, SIGKDD, SIGARCH and others. The SIGEVO Best Dissertation Award will be given annually to a maximum of 1 winner and a maximum of 2 honorable mentions. The award presentation will take place at the awards ceremony of GECCO. The award will carry a monetary value, contributed by SIGEVO, of $2,000 to be awarded to the winner and $1,000 to each of the honorable mentions. The award winner and honorable mentions will each receive a beautiful plaque.

Dissertations are reviewed by a selection committee for technical depth and significance of the research contribution, potential impact on the field of evolutionary computing, and quality of presentation. This year, the members of the selection committee were:

Members of the Selection Committee

The committee received and reviewed 9 nominations this year, in topics as diverse as multi-objective optimization, automated software engineering, grammatical evolution, quality-diversity algorithms, hyper-heuristics, and runtime analysis

The winner of the 2022 SIGEVO Best Dissertation Award is Automatic Algorithm Configuration: Methods and Applications by Marcelo De Souza from Universidade Federal do Rio Grande do Sul (Brazil) with the following citation from the selection committee: “This dissertation takes a deep dive on the topic of automatic algorithm configuration, with a particular focus on heuristic and evolutionary algorithms. It presents novel methods to improve the automatic configuration of algorithms, producing better results, reducing the computational effort, and facilitating its analysis and understanding. It also proposes a solver that applies such techniques to automatically design heuristic and evolutionary algorithms for a broad class of binary optimization problems.”

Figure 6.3 from “Automatic Algorithm Configuration: Methods and Applications” by M. De Souza: Test results after configuring ACOTSP with unrepresentative training instances. Instances with random distances (starting with ‘r’ and in blue) were used only for testing, while Euclidean instances (in black) were used for both training and testing.

In addition to the winner, the committee decided to award two honorary mentions due to their high quality.

The first honorary mention is given to Algorithm Configuration Landscapes: Analysis and Exploitation by Yasha Pushak from the University of British Columbia (Canada) with the following citation: “This dissertation challenges long-held beliefs in and brings attention to a mostly overlooked research area, namely search landscape analysis for automated algorithm configuration (AAC) and AutoML. It shows that these landscapes have a simple structure and contain patterns that can be exploited to improve the state of the art in automated algorithm configuration. This result is demonstrated on various applications of AAC.”

Figure 6.1 from “Algorithm Configuration Landscapes: Analysis and Exploitation” by Y. Pushak: Four examples of one-dimensional parameter response slices. From left to right and top to bottom: EAX’s Npop on the TSP-RUE-1000-3000 instance set, CaDiCaL’s keepglue on the circuit-fuzz instance set, LKH’s BACKBONE_TRIALS on the TSP-RUE-1000-3000 instance set and CPLEX’s mip_limits_submipnodelim on the Regions200 instance set.

The second honorary mention was given to Enhancing Evolutionary Algorithm Performance with Knowledge Transfer and Asynchronous Parallelism by Eric O. Scott from George Mason University (USA) with the following citation: “This dissertation advances the theory of evolutionary knowledge transfer and demonstrates the effectiveness of asynchronous evolutionary algorithms. It provides guidelines on how to design asynchronous evolutionary algorithms, proves no-free lunch theorems and runtime results for evolutionary knowledge transfer methods and considers representation-based versions of evolutionary knowledge transfer. Thus, it perfectly combines rigorous theoretical research with empirical analysis. The techniques proposed here solve important problems in scaling optimization to large compute clusters and generalizing intelligently to new tasks. The resulting software is widely used and has been tested on one of the world’s largest supercomputers”.

Figure 5.2 from “Enhancing Evolutionary Algorithm Performance with Knowledge Transfer and Asynchronous Parallelism” by Eric O. Scott: An example of how a CGP genome is decoded into a graph structure, envisioned here as a logic circuit.

The next edition of the award will welcome dissertations defended between January and December of 2023. The deadline for submitting nominations is March 1st, 2024. More information is available at: https://sig.sigevo.org/index.html/SIGEVO+Dissertation+Award

We encourage you to disseminate this information among your colleagues and students! We are looking forward to your nominations


ACM TELO Indexed by Scopus

We are pleased to report that Scopus has agreed to index ACM Transactions on Evolutionary Learning and Optimization (TELO). Only about 25% of applications for inclusion in the index are successful. The fact that ACM TELO was accepted on the first attempt is a strong confirmation of its high quality and will help to further enhance its visibility and reputation.

The latest issue of ACM TELO: The Best of GECCO 2022, Part I

Volume 2, Issue 4. December 2022

EditorsJuergen BrankeManuel López-Ibáñez

Publisher: Association for Computing Machinery, New York, US

Table of Contents

  • Editorial to the “Best of GECCO 2022” Special Issue: Part I. Jonathan Fieldsend, Markus Wagner. https://doi.org/10.1145/3606034
  • Crossover for Cardinality Constrained Optimization. Tobias Friedrich,Timo Kötzing, Aishwarya Radhakrishnan, Leon Schiller, Martin Schirneck, Georg Tennigkeit, Simon Wietheger. https://doi.org/10.1145/3603629
  • Online Damage Recovery for Physical Robots with Hierarchical Quality-Diversity. Maxime Allard, Simón C. Smith, Konstantinos Chatzilygeroudis, Bryan Lim, Antoine Cully. https://doi.org/10.1145/3596912
  • Transformation-Interaction-Rational Representation for Symbolic Regression: A Detailed Analysis of SRBench Results. Fabrício Olivetti De França. https://doi.org/10.1145/3597312
  • Covariance Matrix Adaptation Evolutionary Strategy with Worst-Case Ranking Approximation for Min–Max Optimization and Its Application to Berthing Control Tasks. Atsuhiro Miyagi, Yoshiki Miyauchi, Atsuo Maki, Kazuto Fukuchi, Jun Sakuma, Youhei Akimoto. https://doi.org/10.1145/3603716

STNWeb: A new visualization tool for analyzing optimization algorithms

Camilo Chacón Sartori & Christian Blum, Artificial Intelligence Research Institute (IIIA-CSIC), Campus of the UAB, Bellaterra, Spain

Gabriela Ochoa, University of Stirling, Stirling, UK

STNWeb (https://www.stn-analytics.com/) is a new web tool for the visualization of the behavior of optimization algorithms such as metaheuristics. It allows for the graphical analysis of multiple runs of multiple algorithms on the same problem instance and, in this way, it facilitates the understanding of algorithm behavior. It may help, for example, in identifying the reasons for a rather low algorithm performance. This, in turn, can help the algorithm designer to change the algorithm in order to improve its performance. STNWeb is designed to be user-friendly. Moreover, it is offered for free to the research community. An article summarizing STNWeb has been recently published in the Software Impacts Journal.

Interacting with STNWeb only requires three simple steps. First, separate data files containing the trajectory data of the considered algorithms must be generated. In the second step, the configuration of the algorithm comparison must be completed on STNWeb, and the data files must be uploaded. Finally, in the third step, a visualization is generated and can be downloaded in PDF format.

Call for Papers

EvoStar 2024

Evostar 2024 conferences are being held in Aberystwyth, Wales, United Kingdom from 3 to 5 April 2024 in hybrid mode. The venue is the Department of Computer Science on Penglais Campus at Aberystwyth University. Read more about Evostar here.

Call for papers (PDF)

Conferences Webpages

Important Dates

  • Submission deadline: 1st November 2023
  • Notification to authors: TBD
  • EvoStar Conference: 3-5 April 2024

The 13th International Workshop on Genetic Improvement

Co-located with the 46th IEEE/ACM
International Conference on Software Engineering,
ICSE 2024, in Lisbon, Portugal
14–20 Apr 2024 (one day)

GI 2024 invites submissions that discuss recent developments in all research areas and applications of Genetic Improvement. GI is the premier workshop in the field and provides an opportunity for researchers interested in automated program repair and software optimization to disseminate their work, exchange ideas and discover new research directions.

Important dates

  • Submission: 9 Nov 2023 (Thu)
  • Notification: 21 Dec 2023 (Thu)
  • Camera-ready: 25 Jan 2024 (Thu)
  • Workshop: 14–20 Apr 2024 (one day)

ACM TELO Special Issues

Learning and Intelligent Optimisation

  • Guest Editors
    • Meinolf Sellmann, InsideOpt, USA
    • Kevin Tierney, Bielefeld University, Germany
  • Submission deadline: 1st December 2023
  • Call for papers (PDF)

Large Scale Optimization and Learning

  • Guest Editors
    • Mohammad Nabi Omidvar, University of Leeds, UK
    • Yuan Sun, La Trobe University, Australia
    • Xiaodong Li, RMIT University, Australia
    • Christian Blum, Artificial Intelligence Research Institute, Spain
  • Submission deadline extended: 1st December 2023
  • Call for papers (PDF)

About this Newsletter

SIGEVOlution is the newsletter of SIGEVO, the ACM Special Interest Group on Genetic and Evolutionary Computation. To join SIGEVO, please follow this link: [WWW].
We solicit contributions in the following categories:

Art: Are you working with Evolutionary Art? We are always looking for nice evolutionary art for the cover page of the newsletter.

Short surveys and position papers. We invite short surveys and position papers in EC and EC-related areas. We are also interested in applications of EC technologies that have solved interesting and important problems.

Software. Are you a developer of a piece of EC software, and wish to tell us about it? Then send us a short summary or a short tutorial of your software.

Lost Gems. Did you read an interesting EC paper that, in your opinion, did not receive enough attention or should be rediscovered? Then send us a page about it.

Dissertations. We invite short summaries, around a page, of theses in EC-related areas that have been recently discussed and are available online.

Meetings Reports. Did you participate in an interesting EC-related event? Would you be willing to tell us about it? Then send us a summary of the event.

Forthcoming Events. If you have an EC event you wish to announce, this is the place.

News and Announcements. Is there anything you wish to announce, such as an employment vacancy? This is the place.

Letters. If you want to ask or say something to SIGEVO members, please write us a letter!

Suggestions. If you have a suggestion about how to improve the newsletter, please send us an email.

Contributions will be reviewed by members of the newsletter board. We accept contributions in plain text, MS Word, or Latex, but do not forget to send your sources and images.

Enquiries about submissions and contributions can be emailed to gabriela.ochoa@stir.ac.uk
All the issues of SIGEVOlution are also available online at: www.sigevolution.org

Notice to contributing authors to SIG newsletters
As a contributing author, you retain the copyright to your article. ACM will refer all requests for republication directly to you.

By submitting your article for distribution in any newsletter of the ACM Special Interest Groups listed below, you hereby grant to ACM the following non-exclusive, perpetual, worldwide rights:

  • to publish your work online or in print on the condition of acceptance by the editor
  • to include the article in the ACM Digital Library and in any Digital Library-related services
  • to allow users to make a personal copy of the article for noncommercial, educational, or research purposes
  • to upload your video and other supplemental material to the ACM Digital Library, the ACM YouTube channel, and the SIG newsletter site

If third-party materials were used in your published work, supplemental material, or video, make sure that you have the necessary permissions to use those third-party materials in your work

Editor: Gabriela Ochoa

Sub-editor: James McDermott

Associate Editors: Emma Hart, Bill Langdon, Una-May O’Reilly, Nadarajen Veerapen, and Darrell Whitley