Newsletter of the ACM Special Interest Group on Genetic and Evolutionary Computation
Editorial
Welcome to the 3rd (Fall/Autumn) 2025 issue of the SIGEvolution newsletter! Our opening article and accompanying cover illustrate evolved static images that appear to move (motion illusions). We continue with a summary of the 2025 Human-Competitive Results Awards (“Humies”) and follow with a PhD dissertation report on benchmarking optimization heuristics.
This issue also shares memoirs of Gustavo Olague (1969-2025) from some of his PhD students, as well as close friends and collaborators.
We conclude with the usual announcements, forthcoming events and call for submissions. Remember to get in touch if you’d like to contribute or have suggestions for future issues of the newsletter.
Gabriela Ochoa (Editor)
Table of Contents
Research Article
Evolutionary Generation of Visual Motion Illusions
Lana Sinapayen, Sony Computer Science Laboratories, National Institute for Basic Biology, Japan.
Abstract
Visual illusions are a popular subject for entertainment, and they are also an active yet still mysterious research subject. There is no consensus on the mechanisms for many of the most popular illusions, and new types of striking visual illusions regularly “go viral” in social media.
This contribution summarizes our research article, where we try to answer the question of why some static images appear to move (motion illusions) through the use of a prediction-based perception model coupled to a simple pattern generation algorithm. Our results support the hypothesis that illusory motion might be the result of perceiving the brain’s own visual predictions, rather than perceiving raw visual input from the eyes. The philosophical motivation of this paper is to call attention to the untapped potential of “motivated failures”, ways for artificial intelligence systems to fail as biological systems fail, as a worthy outlet for understanding the brain.
Introduction
In a previous paper, “PredNet”, a neural network trained to predict video frames from previous video input (predict t+1 from t), was shown to be sensitive to motion illusions: when presented with specific static images known to “work” on humans, for example the famous Rotating Snakes illusion, the network predicted the same type of motion that humans erroneously perceive in the image. You perceive the static circles as if they were rotating, and so does PredNet. Conversely, when the illusion is broken by changing the pattern, both humans and PredNet stop detecting motion. In this paper, we wondered if PredNet could be used as a judge to classify images generated by another network (a model called CPPN: Compositional Pattern-Producing Network).

An example of motion prediction by PredNet. The red lines are the motion vectors (amplified for clarity) representing the direction and amplitude of the motion predicted by the network; in this case, a strong rotational motion.
Approach
The CPPN outputs an image based on a set of parameters. PredNet gives a score to this image: 0 for non-illusions, 1 for extremely strong illusions, and everything in between. CPPN parameters (its “genes”) that produced highly rated images are recorded, modified at random (like random mutations from biological evolution), and the images produced by these new parameters are scored again by PredNet (the “fitness”). We call this system Evolutionary Illusion Generation (EIGen).

Results
After several generations of this genetic algorithm, we obtain images that are consistently highly ranked by PredNet. When showing these images to people informally, we learned that they did perceive motion in the images: our algorithm had seemingly produced convincing motion illusions. In yet unpublished results, we formally tested the images with human participants, and found that indeed, the artificially produced images work as motion illusions. So not only can PredNet detect illusions in images created by humans, but humans can detect illusions in images created using PredNet. This strongly suggests commonalities between the mechanism of illusory motion between PredNet and humans. And since PredNet was solely trained to predict video frames, it suggests that human perception of illusions is also due to the predictive abilities of our brain.
You can read a version of the paper here (the version with human data is in the process of being published) and generate illusions using EIGen here. The code is available here.
About the Author

Lana Sinapayen is a French researcher based in Japan, working on various projects related to Artificial Life, AI, and Astrobiology. She is an associate researcher at Sony Computer Science Laboratories and an associate professor at the National Institute for Basic Biology in Japan. She is also the Research Chair of the International Society for Artificial Life’s (ISAL) board of directors and a member of ISAL’s DEI committee.
Event Overview
22nd Human-Competitive Results Awards (“Humies”), 2025
Erik Goodman, Michigan State University, United states
The GECCO 2025 Conference was held July 14-18 this year, once again in hybrid mode. Nearly 600 people came to Málaga, Spain to attend in person, and an additional 200 or so registered to attend virtually. The finals of the Humies competition (see www.human-competitive.org for the Call for Entries) was a plenary session on Thursday, July 17, that was attended by about 150 people. The eight finalists earlier selected by the judges (from among 15 entries this year, and based on the papers and entry forms they submitted) 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 humans but are generated automatically by computers. The judges viewed all presentations and then held their deliberations to select the prize winners. All 15 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 saw a tie for the Bronze Award. The Gold Award included US$5,000; the Silver Award $3,000, and the Bronze Awards, $1,000 each. The judges for the competition were Wolfgang Banzhaf, Stephanie Forrest, Erik Goodman, Lee Spector and Darrell Whitley. Publicity for the Humies was done by Bill Langdon. On-line attendance at the presentation session was 108 persons, and about one hundred attended in person.
A link to the video of the entire Humies session, hosted by Whova, which handled the on-line participation, is here for those interested in watching. As usual, after considering all of 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, with a $5,000 prize, was awarded to the team of Elliot Meyerson, Olivier Francon, Darren Sargent, Babak Hodjat, and Risto Miikkulainen, all from Cognizant AI Lab in San Francisco. The oral presenter, Risto Miikkulainen, is also a professor at the University of Texas, Austin. Their paper was entitled “Unlocking the Potential of Global Human Expertise,” and was published in Advances in Neural Information Processing Systems (NeurIPS), vol. 37, pp. 119227–119259
They created an evolutionary AI framework entitled “RHEA,” which, in their words, “fills this role by distilling knowledge from diverse models created by human experts into equivalent neural networks, which are then recombined and refined in a population-based search. The framework was implemented in a formal synthetic domain, demonstrating that it is transparent and systematic. It was then applied to the results of the XPRIZE Pandemic Response Challenge, in which over 100 teams of experts across 23 countries submitted models based on diverse methodologies to predict COVID-19 cases and suggest non-pharmaceutical intervention policies for 235 nations, states, and regions across the globe. Building upon this expert knowledge, by recombining and refining the 169 resulting policy suggestion models, RHEA discovered a broader and more effective set of policies than either AI or human experts alone, as evaluated based on real-world data. The results thus suggest that AI can play a crucial role in realizing the potential of human expertise in global problem-solving.”
Silver Award

The Silver Award and $3,000 went to a team which included Niki van Stein (who presented), Thomas Bäck, Haoran Yin, and Anna V. Kononova, all from Leiden University in the Netherlands. Their paper is entitled, “LLaMEA: A Large Language Model Evolutionary Algorithm for Automatically Generating Metaheuristics.” It was published in IEEE Transactions on Evolutionary Computation, vol. 29, no. 2, pp. 331-345, April 2025.
They submitted a second paper demonstrating the application of LLaMEA, entitled “Optimizing Photonic Structures with Large Language Model Driven Algorithm Discovery,” published in GECCO Companion Proceedings, 2025. LLaMEA is a novel LLM evolutionary algorithm framework that leverages GPT models “for the automated generation and refinement of algorithms. Given a set of criteria and a task definition (the search space), LLaMEA iteratively generates, mutates, and selects algorithms based on performance metrics and feedback from runtime evaluations. This framework offers a unique approach to generating optimized algorithms without requiring extensive prior expertise.” They show how this framework can be used to generate automatically novel closed box metaheuristic optimization algorithms for box-constrained, continuous optimization problems. The abstract also asserts that “LLaMEA generates multiple algorithms that outperform state-of-the-art optimization algorithms (covariance matrix adaptation evolution strategy and differential evolution) on the 5-D closed box optimization benchmark (BBOB). The algorithms also show competitive performance on the 10- and 20-D instances of the test functions, although they have not seen such instances during the automated generation process. The results demonstrate the feasibility of the framework and identify future directions for automated generation and optimization of algorithms via LLMs.” The second paper applies this framework to a photonics problem to automatically discover optimization algorithms for the design of photonic structures. Their results compared well against established methods in photonic inverse design.
Bronze Award (Tie)

One Bronze Award and $1,000 went to Marcos Hernández-Gil, Ángel Ramos-de-Miguel, David Greiner, Domingo Benítez, Ángel Ramos-Macías, and José María Escobar, all of Universidad de Las Palmas de Gran Canaria, Spain. Their paper was entitled, “A computational model for multiobjective optimization of multipolar stimulation in cochlear implants: An enhanced focusing approach,” and appeared in Expert Systems with Applications, Vol. 280, 2025, Article 127472.
As they describe in their abstract, “Multipolar stimulation has been demonstrated to improve auditory perception in individuals with cochlear implants by generating more focused electric fields through simultaneous activation of multiple electrodes. In this study, we propose a novel approach to multipolar stimulation that aims to achieve the narrowest possible pattern of current densities at target neurons. Our goal is to find the optimal profile of currents delivered by the electrodes that maximizes the focusing for a specific power consumption, or alternatively, which minimizes the power for a given focusing. To this end, we have designed two objective functions which are optimized through multiobjective evolutionary algorithms. These objective functions are evaluated using a patient-specific finite element volume conduction model that replicates the cochlear geometry and electrical behavior of the implant. Experimental results demonstrate that this approach achieves tighter current density focusing compared to phased-array stimulation, albeit with higher power consumption. Additionally, it is possible to reach non-dominated solutions that simultaneously improve the focusing and power consumption of both monopolar and phased-array stimulation.”
Bronze Award (Tie)

Another Bronze Award and $1,000 went to the team of Nina Gubina, Andrei Dmitrenko, Gleb Solovev, Lyubov Yamshchikova, Oleg Petrov, Ivan Lebedev, Nikita Serov, Grigorii Kirgizov, Nikolay Nikitin, Vladimir Vinogradov, all but one of whom is at ITMO University, St. Petersburg, Russia. Their paper is entitled “Hybrid Generative AI for De Novo Design of Co-Crystals with Enhanced Tabletability” and appeared in NeurIPS, Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024.
As their abstract describes it, “Co-crystallization is an accessible way to control physicochemical characteristics of organic crystals, which finds many biomedical applications. In this work, we present Generative Method for Co-crystal Design (GEMCODE), a novel pipeline for automated co-crystal screening based on the hybridization of deep generative models and evolutionary optimization for broader exploration of the target chemical space. GEMCODE enables fast de novo co-crystal design with target tabletability profiles, which is crucial for the development of pharmaceuticals. With a series of experimental studies highlighting validation and discovery cases, we show that GEMCODE is effective even under realistic computational constraints. Furthermore, we explore the potential of language models in generating co-crystals. Finally, we present numerous previously unknown co-crystals predicted by GEMCODE and discuss its potential in accelerating drug development.”
More Information
See any of the finalists’ presentation materials, papers and entry forms at www.human-competitive.org. And start preparing your own best human-competitive work for the 2026 Humies competition at GECCO next July, in person in San Juan, Costa Rica, or virtually! Any questions? Contact Erik Goodman, goodman at msu.edu. Humies photo credits: Wolfgang Banzhaf.
PhD Dissertation Report
From Benchmarking Optimization Heuristics to Dynamic Algorithm Configuration
Diederick Vermetten, Leiden University, The Netherlands

Benchmarking plays an important role in understanding the strengths and weaknesses of optimization heuristics. However, data collected from one benchmarking study is not limited in its use to answering that single original research question. By sharing and re-using data, we can gain new insights into, for example, the complementarity of different algorithms. In this thesis, we start from the creation of a software framework which supports the benchmarking process, and utilize the resulting data to investigate meta-algorithmic questions, starting from algorithm configuration and moving to dynamic algorithm selection: switching between optimizers during the search process.
Benchmarking: IOHprofiler
It is important that robust benchmarking pipelines are easily accessible. Within the optimization community, several benchmarking frameworks have been introduced with this goal in mind. In this thesis, we focused on the IOHprofiler framework. Using a modular design structure, IOHprofiler can be integrated with many other commonly used tools for black-box optimization and benchmarking. These integrations form the backbone for the experiments used throughout this thesis, as the corresponding performance and algorithm behavior data can be easily reused. This way, we can go beyond the common ‘competitive’ benchmarking practice, where we only care about the algorithm with the best average performance, to gaining insights about the complementarity between different algorithms

Modular Algorithm Design
One of the key ways in which algorithm complementarity can be beneficial is in the context of algorithm selection and configuration. Instead of relying on a single algorithm for a wide set of problems, we use a set of features to determine which algorithm or algorithm configuration to use. This meta-learning task benefits from variety between algorithms, which can be achieved by looking at different types of algorithms, but also within an algorithm family by considering a wide range of modifications proposed over the last decades. We show how two modularizations of popular evolutionary algorithms, CMA-ES and Differential Evolution, can lead to insights about combinations of algorithmic ideas, resulting in improved performance over previous hand-designed versions of these same algorithms.
Switching Between Optimization Heuristics
While algorithm complementarity exists on the problem level, looking at benchmark data from a variety of sources also reveals a complementarity in performance within individual functions. While some algorithms are great at finding promising regions, others excel at fast convergence once this region is found. As such, the notion of dynamic algorithm selection (DynAS), where we can switch between different algorithms during the optimization procedure, is studied in detail. We highlight the inherent potential in this approach, while simultaneously investigating which aspects of this switching procedure need to be further developed to create reliably dynamic algorithm combinations.
Benchmarking Dynamic Algorithm Selection

Gallagher101 (right).
The difficulty of finding the performance potential of DynAS is due in part to the difficulty of creating reliable benchmarking setups for this scenario. While collections of problems for benchmarking optimization algorithms are widely available, they are not necessarily well-suited for meta-learning scenarios, as there is no truly fair way to create the commonly required train-test set distinction used by the required machine learning techniques. For this reason, we propose a benchmark problem generator based on the commonly used BBOB problem suite, which can be used to generate arbitrary amounts of training and testing problems to benchmark these meta-learning mechanisms.
About the Author

Diederick Vermetten is a post-doctoral researcher at Sorbonne Université, CNRS, LIP6, where he works on the ‘DynaBBO’ project. He obtained his PhD with distinction from LIACS, Leiden University in February 2025, where he was supervised by Thomas Bӓck, Hao Wang and Carola Doerr. He is one of the core developers of the IOHprofiler project, and his main research interests include benchmarking, algorithm selection/configuration and dynamic algorithm selection.
Link
Full dissertation (open access)
Obituary
Remembering Gustavo Olague (1969 – 2025)
Aarón Barrera Román, Axel Martínez Navarro, Eddie Clemente, León Dozal, Enrique Dunn, Daniel Hernández, Benjamín Hernández Valencia, Rocio Ochoa, Cynthia Pérez , José Antonio Pérez, Cesar Puente, Leonardo Trujillo, Juan Villegas Cortez

It saddens us to share some unfortunate news, our good friend and colleague, Gustavo Olague, passed away suddenly this past July 26th, in Ensenada, Mexico, close to CICESE where he had worked as a researcher since 1999, and reached the highest level of scientific recognition within the Mexican system. He is survived by his wife and children, one of whom, Matthieu Olague, was active in research work with him over recent years. This is an irreparable loss for his family and friends, and for all the scientific community, especially in Mexico, where he was a great mentor and teacher. We are fortunate and proud to have had him as our doctoral advisor, and more importantly to consider him our friend.
Gustavo was born in Chihuahua, Mexico, in 1969. He obtained his Ph.D. in Computer Vision, Graphics and Robotics from the Institut Polytechnique de Grenoble (INPG) and the Institut National de Recherche en Informatique et Automatique (INRIA), where he worked with Roger Mohr. We are convinced that his scientific contributions to our community, particularly in the field of Evolutionary Computer Vision (the title of his book on the subject), will have a lasting impact on those of us who work at the intersection of these two areas of Artificial Intelligence. We firmly believe that he will be a reference point for future generations. At CICESE, he founded the EvoVision Lab, of which we were all proud members. Together, we explored these and a wide range of other scientific and philosophical questions.
Gustavo was a person of faith, and believed in the central role of purpose, both in human life and in artificial systems. One of the purposes of Gustavo was to teach and help young researchers in Mexico, from all walks of life, find joy and promise in scientific research. He allowed many students to connect with the wonderful world of scientific knowledge and inquiry, a true challenge in our country that he ably navigated.
We hope that you will remember Gustavo fondly, as do we. May he rest in peace.
Francisco Fernandez de Vega
The year 1999 is marked in bold italics on my calendar: it was the year I attended my first international conference on evolutionary algorithms; it was also my first summer as a predoctoral researcher outside Europe; and my first visit to the United States. But above all, at that first historic GECCO conference, held in Orlando, and the series of editions that followed in subsequent years, I had the opportunity to meet those who would become colleagues and friends for life. We were a young and enthusiastic Spanish speakers group with evolutionary algorithms as our banner, when Artificial Intelligence was not yet a trending topic, and we were meeting the pioneers and leaders for the first time: J. Holland, D. Goldberg, J. Koza, E. Goodman… Young people who went on to become colleagues and friends through shared experiences and publications from afar: Iñaki Hidalgo, Hernán Aguirre, Jose A. Lozano, Carlos Cotta, JJ Merelo, Anna Esparcia… and our dear Gustavo Olague, to name but a few.
It was in the summer of 2000, after the second GECCO conference, this time in Las Vegas, where we will always remember our first bets “to win” a measly $50 at the casinos, when Iñaki and I traveled to CICESE in Ensenada, Mexico, to give some talks about our evolutionary battles. That year, 2000, was the first in a series of visits and exchanges that would last two decades, between Gustavo, his students, and our group from Extremadura, Spain. It was a fruitful relationship that, starting in 2016, was put on hold due to the cuts that our colleague suffered from his country’s government. Two decades of summer courses, conferences, stays, and congresses. These stays helped our Spanish children learn to “speak Mexican,” and also helped Matthew, Gustavo’s son, learn to “speak Spanish.” What good times we had with Gustavo, his wife Carmela, and Matthew at those harvest festivals in the Guadalupe Valley, at La Primera de Ensenada, at his El Santo film series, or blues festival—a city full of culture in the summer—the birthplace of the “margarita,” sharing tacos and burritos, visiting the Colorado Canyon… Years of shared joy with his closest doctoral students: Leonardo, Eddie, Daniel, Cynthia, and many others of his collaborators, such as Juan, Humberto,…
Those were years of many ideas, conceived in the limelight of Mexican or Spanish bars, washed down with Coronitas or Negra Modelo on the shores of the Pacific, or Mahou and Cruzcampos in Mediterranean lands. There were many scientific joys, joint articles, family gatherings, meetings with students, trips—to Santiago de Compostela or Mexico City—and also, life itself, shared difficulties and problems.
Gustavo always tried to help young students from his homeland pursue a scientific dream. A dream he lived firsthand with his doctorate in Grenoble, France, under the supervision of Roger Mohr, his PhD advisor, a direct descendant of Descartes, and which, back in Mexico, often clashed with a difficult reality. A dream he never gave up on: science above publications; ideas beyond citations; friends before scientific or social relevance; serious work despite the funds received. His serious work earned him international recognition, including the Talbert Abrams Award from the American Society for Photogrammetry and Remote Sensing (ASPRS) and bronze medals at the Humies Awards in 2006 and 2009.
Gustavo brightened up any gathering and was a staunch defender of his ideas, so much so that he would try to convince anyone willing to engage in sometimes endless discussions, such as the last one I remember at the 2019 Genetic Programming Theory and Practice event at MSU—the last time we met. But who doesn’t have flaws!
Despite the distance, we were able to share a lot, including our faith-based outlook of life. He called me brother and friend. And over the years, “Amigos para siempre” became the celebratory song for the gala dinners at the European EVOSTAR and the Spanish MAEB evolutionary computation conferences.
Last Thursday, August 24, we had a video meeting to resume and relaunch plans for the future. Plans that were abruptly and unexpectedly interrupted on Saturday, August 26.
It is a good time to return to this inescapable “Amigos para siempre”: this Spanish title refers to a “forever” that sooner or later we must face with faith, so necessary for those who believe in an infinite material universe in time, without cause or origin, and in which, ultimately, the time of human history will tend toward nothingness, and will dissolve like sugar in water; or a different kind of “forever” for those of us who look with faith the transition to the next life, and believe in the “I am who I am” that our parents passed on to us, and whose circumstantial evidence reaffirms us, and allows us to understand that this “friends forever” has a real and concrete meaning, and sooner than later, Gustavo, we will see each other again.
R.I.P.
Evelyne Lutton
I first met Gustavo at INRIA, during his PhD with Roger Mohr. He came to see me because I was then working, together with Marc Schoenauer, to bring together the French community around Artificial Evolution. From the very beginning, he stood out for his curiosity and his generosity of spirit. He forged a lasting bridge between Mexico and France, a bridge made of doctoral students, postdoctoral fellows, and countless research visits over the years. Pierrick Legrand was one of these post-docs and Leonardo Trujillo one of these PhDs. In this role of connecting people, Gustavo also wove another bridge — both scientific and deeply friendly — with Spain, a bridge through which I had the joy of meeting Francisco Fernández de Vega.
We would meet regularly at the conferences of the evolutionary computation community, GECCO, PPSN, EuroGP, and many others. Beyond his brilliant individual results, Gustavo played a leading role in shaping evolutionary computer vision into a recognised research field, co-editing landmark special issues and surveys that crystallised the area and gave it recognition.
Together, we co-authored a series of papers of which I am proud, reflecting a shared aspiration: to use the principles of evolution and artificial life to make vision systems more adaptive, creative, and intelligent. We explored how evolutionary algorithms could discover image operators, descriptors, and behaviours, opening the way to advances in texture segmentation, object recognition, and place recognition. Our common work is more than the sum of its parts: it tells the story of a beautiful collaboration, guided by curiosity and by the conviction that the theory of evolution is a remarkably rich lens for developing artificial intelligence and perception.
Gustavo’s personality was dynamic and joyful. Scientific discussions with him were always profound, funny, original, imaginative, engaging, enthusiastic, and always nourished by a wide-ranging culture. With him, research was not only fruitful, but also inspiring and stimulating.
I feel privileged to have met Gustavo during my career, as a colleague and as a friend. It is hard to imagine that he has left the scientific stage so suddenly. He leaves us saddened and dismayed. I will miss him deeply.
Pierrick Legrand

I had the privilege of meeting Gustavo Olague during my postdoctoral fellowship at the CISESE Research Center in Ensenada in 2005. At that time, he was leading the research team that welcomed me.
Allow me to share with you a few personal anecdotes that I had the honor of experiencing with Gustavo.
The first one concerns my arrival in Ensenada. The journey from France to San Diego had been quite long, and Gustavo kindly arranged for a driver to pick me up at the San Diego airport and take me to the laboratory. I believe it was around midnight when I finally arrived at CISESE, and there was Gustavo, waiting for me. He welcomed me with open arms, literally. Although I was exhausted, he had prepared a tour of the laboratory, a visit that I undertook wholeheartedly, even in the middle of the night. He had organized demonstrations, and I was deeply impressed to discover what he and his doctoral students and colleagues had achieved together.
At the end of the visit, he led me up to the top of the CISESE site, on the summit of the hill, to show me the accommodation he had arranged for me. It was a mobile home shared with other foreign researchers and doctoral students. Gustavo had even asked them to prepare a meal so that I would feel properly welcomed. Thus, my stay in Mexico began under the best possible auspices. This was one of the many lessons I received from Gustavo, and from the Mexican people more broadly: always strive to give the warmest welcome to newcomers.
During my stay, Gustavo and I worked closely together. My role was to train the Mexican team in fractal analysis, and in return I had the opportunity to deepen my knowledge of artificial evolution under the benevolent guidance of Gustavo and Evelyne Lutton. All those who had the privilege of working with him could attest to his scientific expertise and his remarkable writing skills. This experience was profoundly formative for me; I learned so much from Gustavo. Yet our exchanges extended well beyond science. Gustavo was a man of deep faith, and we often spoke at length about the role of religion and family in our lives. He was an ideal partner in dialogue, thoughtful, respectful, and always ready to engage in calm and meaningful discussions, even when our views diverged profoundly.
Earlier, I spoke of hospitality, and of how Gustavo taught me the importance of caring for newcomers. I had the honor of being invited several times to Gustavo’s home, where I met his family and shared barbecues with them. It was on one such occasion that I saw the Super Bowl for the very first time. I was also invited to his son’s birthday celebration, held at a recreational center with video games, where I felt like part of the family (Gustavo even gave me tokens so that I could play alongside the children). Among the many activities he so generously organized, I especially remember the magnificent visit to La Bufadora in Ensenada. Throughout my entire stay, Gustavo sacrificed his time solely for my well-being, and I will never forget that.
In conclusion, I would say that Gustavo Olague was not only a driving force in the professional journeys of many of us, but also a profound influence on my own path in life.
Journal Latest Issues


Genetic Programming and Evolvable Machines (GPEM)
Editor-in-chief: Leonardo Trujillo
Volume 26, Issue 2
December 2025

ACM Transactions on Evolutionary Learning and Optimization (TELO)
Editors-in-chief: Jürgen Branke, Manuel López-Ibáñez
Volume 5, Issue 3
September 2025
Forthcoming Events

International Conference on Evolutionary Computation Theory and Applications (ECTA)
ECTA is part of IJCCI, the 17th International Joint Conference on Computational Intelligence.
Registration to ECTA allows free access to all other IJCCI conferences.
IJCCI 2025 will be held in conjunction with ICINCO 2025, WEBIST 2025, IN4PL 2025, IC3K 2025 and CoopIS 2025. Registration to IJCCI allows free access to the ICINCO, WEBIST, IN4PL, IC3K and CoopIS conferences (as a non-speaker).
Conference Areas
1. Theory and Methods
2. Applications
Conference Chair: Joaquim Filipe, Polytechnic Institute of Setubal / INSTICC, Portugal
Program Chair: Niki van Stein, Leiden University, Netherlands
Keynote Speakers
Luís Paulo Reis, University of Porto, Portugal
Pietro Ducange, University of Pisa, Italy
Fei Liu, City University of Hong Kong, Hong Kong
Henry Prakken, Utrecht University, Netherlands
Call for Submissions

EvoStar 2026
The EvoStar conferences are being held in Toulouse, France from 8 to 10 April 2026 in hybrid mode. The four conferences are:
- EuroGP 29th European Conference on Genetic Programming
- EvoApplications 29th European Conference on the Applications of Evolutionary and Bio-inspired Computation
- EvoCOP 26th European Conference on Evolutionary Computation in Combinatorial Optimisation
- EvoMUSART 15th International Conference (and 20th European event) on Evolutionary and Biologically Inspired Music, Sound, Art and Design
Important dates
- Submission deadline: 1 November 2025
- Notification to authors: 10 January 2026
- Camera-ready submission: 24 January 2026
- Author’s mandatory registration: 11 February 2026
- Early registration deadline: 4 March 2026
- Late-Breaking Abstracts submission: 29 March 2026

Generative AI and Evolutionary Computation for Software Engineering
Generative models, and mainly large language models, are already widely used tools in real-world software development. They assist with writing code, generating tests, fixing bugs, and more. While these tools are powerful and quite useful, they still struggle with reliability, maintainability, and meeting complex requirements. This is where evolutionary algorithms, and especially genetic programming techniques, can make a decisive contribution. Unlike generative models that primarily rely on learned patterns, evolutionary algorithms offer a search-based approach to systematically explore the solution spaces. By combining the strengths of generative models with the flexibility and robustness of search-based techniques, we could build hybrid systems that produce even better and more reliable results.
The focus of this special issue is on the integration of generative methods and evolutionary computation to advance software engineering tasks. It aims to highlight approaches where evolutionary methods enhance, guide, or refine the output of generative models to produce better software solutions
Key Dates
- Submission Deadline: 1 December 2025
- Reviews: 1 March 2026
- Revision Deadline: 15 April 2026
- Final Acceptance Notification: 15 May 2026
Guest Editor
Dominik Sobania, Johannes Gutenberg University, Mainz, Germany ([email protected])
Thematic Area Editor (Software Engineering)
Justyna Petke, University College London, UK
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 [email protected].
All the issues of SIGEVOlution are also available online at: https://evolution.sigevo.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-editors: James McDermott and Nadarajen Veerapen
Associate Editors: Emma Hart, Bill Langdon, Una-May O’Reilly, and Darrell Whitley