#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: a) From Pixels to Metal: AI-Empowered Numismatic Art #2. The name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s): Penousal Machado University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal machado@dei.uc.pt +351 239790016 Tiago Martins University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal tiagofm@dei.uc.pt +351 239790016 João Correia University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal jncor@dei.uc.pt +351 239790016 Luís Espírito Santo University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal lesanto@dei.uc.pt +351 239790016 Nuno Lourenço University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal naml@dei.uc.pt +351 239790016 João Miguel Cunha University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal jmacunha@dei.uc.pt +351 239790016 Sérgio Rebelo University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal srebelo@dei.uc.pt +351 239790016 Pedro Martins University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal pjmm@dei.uc.pt +351 239790016 João Bicker University of Coimbra, CISUC/LASI, DEI Pólo II - Pinhal de Marrocos 3030-290, Coimbra, Portugal bicker@dei.uc.pt +351 239790016 #3. The name of the corresponding author (i.e., the author to whom notices will be sent concerning the competition): Penousal Machado (machado@dei.uc.pt) João Correia (jncor@dei.uc.pt) Nuno Lourenço (naml@dei.uc.pt) #4. The abstract of the paper(s): a) This paper describes our response to a unique challenge presented by the Portuguese National Press-Mint: to use Artificial Intelligence to design a commemorative coin that celebrates the “digital world.” We explain the process of this coin’s co-creation, from conceptualisation to production, highlighting the design process, the underlying rationale, key obstacles encountered, and the technical innovations and developments made to meet the challenge. These include the development of an evolutionary art system guided by Contrastive Language–Image Pre-training (CLIP) and Machine Learning-based aesthetic models, a system for prompt evolution, and a representation for encoding genotypes in mintable format. This collaboration produced a limited edition 10 euro silver proof coin, with a total of 4 000 units minted by the National Press-Mint. The coin was met with enthusiasm, selling out within two months. This work contributes to Computational Creativity, particularly co-creativity, co-design, and digital art, and represents a significant step in using Artificial Intelligence for Numismatics. #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: The work satisfies the following criteria: D, E, F, 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): (D) The result is publishable in its own right as a new scientific result independent of the fact that the result was mechanically created. The result of this research is publishable in its own right as a new scientific result independent of the mechanical creation because it proposes a framework that combines Genetic Programming (GP) for image generation, CLIP [Radford et al., 2021] and Machine Learning (ML) aesthetic models; and an Grammatical Evolution (GE) system for evolutionary prompt design. Such integration represents a technological innovation, expanding the frontiers of AI and computational creativity. Secondly, the application of these AI tools to the field of numismatics, which is historically and culturally rich, is pioneering. This application is not merely a demonstration of technology but a meaningful contribution to the domain of numismatics, where it was used to aid in the design of a commemorative coin that was minted into a physical artefact. The ability to apply AI techniques to analyze, interpret, and generate knowledge in numismatics underscores the interdisciplinary impact and the practical utility of the research. Therefore, the novelty and scientific significance of both the developed AI tools and their application validate the research's publication as an independent scientific result. These contributions advance the state of the art in AI and provide valuable new methods and perspectives to the study of numismatics. (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. In numismatics, AI is primarily used for classifying coins, as evidenced by works such as Anwar et al. [2021]. The recent AI-designed coin introduced by the Pressburg Mint [2023] represents a significant advancement in this area, showcasing the potential of AI in artistic design. However, upon inspection of this AI-designed coin, it is suspected that it may have been produced using off-the-shelf models like DALL-E or Stable Diffusion, which do not exhibit the artistic value of our approach. Our project leverages an Evolutionary Computation (EC) engine that integrates the semantic interpretation capabilities of CLIP [Radford et al., 2021], Machine Learning (ML) aesthetic models, and an EC system for evolutionary prompt design. This innovative combination results in a more sophisticated and nuanced approach to AI-generated coin design, surpassing the capabilities of pre-existing off-the-shelf models. Moreover, while the exact timeline of the Pressburg Mint's coin design relative to our project is unclear, it is evident that our approach, developed independently and without knowledge of their work, provides a more advanced and artistically aligned solution. Our research embodies a deeper understanding of computational creativity, applying advanced AI tools to create aesthetically pleasing designs and infusing them with historical and cultural significance. Our approach shows the application of AI in numismatic design, showcasing a level of innovation and creativity that matches or exceeds the most recent human-created solutions. This advancement is not only in the technical sophistication of the AI tools but also in their meaningful application, setting a new benchmark in the field. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. The Pressburg Mint [2023] introduced a coin designed by AI. No detailed study on the design process is available, yet by inspection of the final product, we suspect it results from an off-the-shelf model such as DALL-E or Stable Diffusion. It is unclear whether this coin was designed before or after our project; however, it was minted before. Our project resulted in the second minted coin in the world but it is the embodiment of co-creation, creativity and art as an output that surpasses the previous attempt and others based on modern generative AI. (G) The result solves a problem of indisputable difficulty in its field. In the era of data-driven generative models, creating new content, not imitation or pastiche, is a hard endeavor in itself. Furthermore, the design of the coin in terms of requirements is in itself a challenge – “We want a coin designed by Artificial Intelligence (AI) about the digital world. Can you do it?”. This initial challenge ignited our collaboration with Portuguese Mint press (INCM) and led us to delve into several fundamental questions: What is a coin? Who is the AI designing the coin for? What is a coin designed by AI? The questions led us to conclude that we are not just generating images to be a part of a face of a coin, we have to build something that enables the AI to design and create - a more anthropomorphic view of AI, as we would have used tools to create the coin, the AI should have and use tools to create one. Moreover, the coin has two sides, which as explained by the INCM team, “both faces of the coin should communicate, convey the same message”. Practical factors like mintability, durability, recognizability, legal regulations, thematic importance, and aesthetic appeal, also created unique challenges for their design. Taking all this into account, for the design of both faces, called respectively the reverse and obverse of the coin, we divided the problem into two guidelines: one side (reverse) should be designed by AI for Humans and the other by Humans for AI (obverse). Available data-driven tools in modern generative AI, while producing impressive results, present challenges in the control of the outputs, which sometimes are biased towards preconceptions that are present on the data that they were trained on. Our approach is not data-driven and uses the tool NEvAr (Genetic Programming based tool by [Machado and Cardoso, 2002]) in an accelerated version via TensorGP [Baeta et al., 2021] for GPU computation to generate the face that is designed by AI. The tool was plugged-in with CLIP and using prompts of MetaPrompter (Grammatical Evolution)[Martins et al., 2023] , NEvAr generated images that would be later selected to be one side of the coin. The whole system solves the problem of imitation and novelty via the interactions of the AI tools to create this face of the coin. The other coin face, the obverse, is a graphical representation of the genetic code of the reverse image. For this side, we created a system that converts the genotype evolved by AI into a visual composition of concentric arcs encoding the mathematical expression corresponding to the reverse image. Continuous arcs represent mathematical functions, while interrupted arcs represent numerical values. It is also important to note that this arc system was designed to be machine-readable, allowing access to the instructions necessary to reproduce the reverse image. Thus, depending on the length of the genotype, we can find obverses with more or fewer arcs. Getting to the final output to answer the requirements was in fact solving some of the limitations and barriers of modern generative AI. Moreover, based on the final coin, we created a unique digital artifact for each of the 4,000 produced coins, illustrating the evolutionary process behind the coin's creation in a 40-second video divided into three segments: the ancestors of the coin's image, the transformation into a 3D coin model, and the creation of a new digital offspring - a new coin face created by continuing the evolutionary process for a few extra generations. The video begins by showcasing the ancestors of the image used on the coin's reverse side, offering a brief overview of the evolutionary process through a genetic morphing animation of selected individuals. Guided by LAION-5B [Schuhmann et al., 2022] to assign fitness, we generated 14,000 coin offspring, sorted them by aesthetic rating, and curated a diverse archive by comparing genotypes and phenotypes, ensuring each of the final 4,000 coins had a unique digital asset. The results also showcase the ability of our approach to generate new suitable artifacts based on a previous output and keep some of the key aspects of its predecessor. It also showcases the flexibility, adaptability and exploratory power that our generative approach provides. #7. A full citation of the paper (that is, author names; publication date; name of journal, conference, technical report, thesis, book, or book chapter; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable) a) @inproceedings{ijcai2024coin, author = {Machado, Penousal and Martins, Tiago and Correia, Jo{\~{a}}o and Esp\'{i}rito Santo, Lu\'{i}s and Nuno Louren{\c{c}}o and Jo{\~{a}}o Cunha and Rebelo, S{\'{e}}rgio and Martins, Pedro and Jo{\~{a}}o Bicker}, booktitle = {Proceedings of the Thrity-Third International Joint Conference on Artificial Intelligence, {IJCAI} 2024, 3-9 August 2024, Jeju, South Korea }, url = {https://www.ijcai.org/proceedings/2024/}, month = {8}, note = {Ai, the Arts and Creativity}, pages = {to appear}, publisher = {International Joint Conferences on Artificial Intelligence Organization}, title = {From Pixels to Metal: AI-Empowered Numismatic Art}, year = {2024}, } #7. 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: Any prize money, if any, is to be divided equally among the co-authors. #8. A statement stating why the authors expect that their entry would be the "best": The overall approach to the problem showcases the co-creativity potential of AI generative tools, the strengths of Evolutionary Machine Learning approaches via the collaboration of Evolutionary Computation systems, namely Genetic Programming and Grammatical Evolution, and Machine Learning. The final result is a physical object – a coin –, an embodiment of the final output of the system which also marks an historical point of AI-Human co-creation. Coins serve as both legal currency and enduring symbols of cultural heritage. Unlike commemorative medals, coins function as both aesthetic objects and practical money. While medals are created to commemorate events, coins have assigned monetary values and facilitate economic transactions under government authorization. Anyone can mint a medal, but only the government has the authority to mint a coin. The dual nature of coins as both functional currency and objects of beauty, combined with practical factors like mintability, durability, recognizability, legal regulations, thematic importance, and aesthetic appeal, creates unique challenges for their design. Considering these requirements, we set out to create a coin that honors the rich tradition of numismatics and the National Press-Mint, while also reflecting our identity as an AI and Design Lab. Our goal was to respect our perspectives on Computational Creativity (CC) research, challenge traditional boundaries, and introduce innovative paths for expression and design. The culmination of this project was the minting of a limited edition (4 000 exemplars) 10 euro silver proof coin by Portuguese Mint press (INCM), which sold out in less than two months at a fixed price of 95 euros each, attesting its success and the public’s approval of the design and inherent design principles. Link to promotional video: https://vimeo.com/913415560 (Portuguese with English subtitles); and post https://cdv.dei.uc.pt/projects/ai-coin . Moreover, we created a unique digital artefact for each of the 4 000 coins, which includes a ``digital offspring'' of the coin. The created digital assets illustrate the evolutionary process that led to the minted coin. Each digital asset is a 40-second video showing the evolutionary process behind the coin's creation, divided into three segments. The video begins by showcasing the ancestors of the image used on the coin's reverse side, offering a brief overview of the evolutionary process through a genetic morphing animation of selected individuals. It then transitions into the second part, where the image morphs into a 3D coin model, symbolizing the minting process. As the 3D coin rotates to display its two faces, the video subtly shifts to the final segment, which features the creation of a digital offspring. This new coin is produced by extending the evolutionary process for several more generations, guided by LAION-5B. We generated 14,000 offspring of the original coin, sorted them by aesthetic rating, and curated them into an initially empty archive, ensuring diversity by comparing genotypes and phenotypes. Only the fittest images were retained, resulting in a unique digital asset for each of the 4,000 coins. The video of the digital assets can be found here: https://vimeo.com/893105678 The results were highlightly acclaimed and received wide recognition at national and international level: https://observador.pt/2023/12/11/casa-da-moeda-lanca-moeda-desenhada-por-inteligencia-artificial-uma-obra-de-arte-que-une-o-biologico-e-o-digital/ https://executivedigest.sapo.pt/noticias/primeira-moeda-desenhada-por-inteligencia-artificial-entra-hoje-em-circulacao-em-portugal/ https://postal.pt/sociedade/veja-a-primeira-moeda-do-mundo-desenhada-por-inteligencia-artificial-em-portugal-que-vale-95e/ https://www.imagensdemarca.pt/artigo/entra-hoje-em-circulacao-em-portugal-uma-moeda-desenhada-por-inteligencia-artificial/ https://www.museudodinheiro.pt/agenda/moedas-e-mentes-um-futuro-cunhado-por-inteligencia-artificial https://www.diariocoimbra.pt/noticia/116991 https://www.e-cultura.pt/evento/37573 #9. 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), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc.: GP (Genetic programming) and Grammatical Evolution (GE) #10. 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: a) August 2024 (in press) Please check "IJCAI-Copyright_Transfer_Agreement.pdf" or "Gmail - Fwd_ Notification - IJCAI-2024 Special Track on AI, the Arts, and Creativity.pdf" #References: [Anwar et al., 2021] - Hafeez Anwar, Saeed Anwar, Sebastian Zambanini, and Fatih Porikli. Deep ancient roman republican coin classification via feature fusion and attention. Pattern Recognition, 114:107871, 2021. [Radford et al., 2021] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agar- wal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision. In ICML, volume 139 of Proceedings of Machine Learning Research, pages 8748–8763. PMLR, 2021. [Machado and Cardoso, 2002] Penousal Machado and Amílcar Cardoso. All the truth about NEvAr. Appl. Intell., 16(2):101–118, 2002. [Baeta et al., 2021] Francisco Baeta, João Correia, Tiago Martins, and Penousal Machado. Tensorgp - genetic programming engine in tensorflow. In Pedro A. Castillo and Juan Luis Jiménez Laredo, editors, Applications of Evolutionary Computation - 24th International Conference,EvoApplications 2021, Held as Part of EvoStar 2021, Virtual Event, April 7-9, 2021, Proceedings, volume 12694 of Lecture Notes in Computer Science, pages 763–778.Springer, 2021. [Martins et al., 2023] Tiago Martins, João Miguel Cunha,João Correia, and Penousal Machado. Towards the evolution of prompts with metaprompter. In EvoMUSART@EvoStar, volume 13988 of Lecture Notes in Computer Science, pages 180–195. Springer, 2023. [Schuhmann et al., 2022] Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. LAION-5B: an open large-scale dataset for training next generation image-text models. In NeurIPS, 2022. [Pressburg Mint, 2023] https://www.einpresswire.com/article/662851149/pressburg-mint-releases-the-world-s-first-silver-coin-designed-by-artificial-intelligence