{"id":1692,"date":"2026-09-01T09:44:30","date_gmt":"2026-09-01T07:44:30","guid":{"rendered":"https:\/\/www.evostar.org\/2027\/?page_id=1692"},"modified":"2026-09-01T09:44:31","modified_gmt":"2026-09-01T07:44:31","slug":"ecxtai","status":"publish","type":"page","link":"https:\/\/www.evostar.org\/2027\/evoapps\/ecxtai\/","title":{"rendered":"Evolutionary Computation for Explainable and Trustworthy AI"},"content":{"rendered":"\n<p>Artificial intelligence (AI) systems are increasingly deployed in settings where predictive accuracy alone is insufficient. Their decisions must also be understandable, reliable, robust, and open to scrutiny. However, many explainability and trustworthiness problems involve competing objectives, complex search spaces, discrete or structured representations, and constraints that are difficult to address with conventional optimization methods.<\/p>\n\n\n\n<p>Evolutionary computation is particularly well suited to these challenges. Population-based search can generate diverse explanations, explicitly balance criteria such as fidelity, stability, diversity, and plausibility, operate with non-differentiable models, and optimize symbolic, rule-based, visual, prototype-based, or counterfactual representations.<\/p>\n\n\n\n<p>The proposed special session will provide a focused forum for researchers developing evolutionary and bio-inspired approaches to the generation, optimization, evaluation, and auditing of explanations and intrinsically interpretable AI systems. It will welcome methodological contributions, evaluation studies, benchmarks, and real-world applications across machine learning, computer vision, image processing, and related areas, with a clear emphasis on the distinctive role of evolutionary search in explainable and trustworthy AI.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">&nbsp;<strong>Scope and Topics&nbsp;<\/strong><\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Evolutionary generation and optimization of local and global explanations&nbsp;<\/li>\n\n\n\n<li>Counterfactual explanations, causal recourse, and actionable explanations&nbsp;<\/li>\n\n\n\n<li>Human-in-the-loop and interactive evolutionary explainability&nbsp;<\/li>\n\n\n\n<li>Genetic programming, symbolic regression, rule learning, and fuzzy systems for interpretable AI&nbsp;<\/li>\n\n\n\n<li>Neuroevolution for transparent, compact, or self-explainable models&nbsp;<\/li>\n\n\n\n<li>Evolutionary feature selection and feature construction for explainability&nbsp;<\/li>\n\n\n\n<li>Prototype-, example-, concept-, and case-based explanations&nbsp;<\/li>\n\n\n\n<li>Evolutionary visual explanations for image processing and computer vision&nbsp;<\/li>\n\n\n\n<li>Quality-diversity methods for generating multiple valid explanations&nbsp;<\/li>\n\n\n\n<li>Evolutionary optimization and analysis of robustness, fairness, uncertainty, and bias&nbsp;<\/li>\n\n\n\n<li>Evolutionary testing and auditing of black-box models and explanation methods\u00a0<\/li>\n\n\n\n<li>Explainable evolutionary reinforcement learning and decision-making\u00a0<\/li>\n\n\n\n<li>Explainability for deep, multimodal, generative, and foundation models\u00a0<\/li>\n\n\n\n<li>Benchmarks, metrics, reproducibility, scalability, and user-centered evaluation\u00a0<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Organisers&nbsp;<\/strong><\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Diego Oliva<\/strong><br>Universidad de Guadalajara, Mexico<br>diego.oliva(at)cucei.udg.mx<\/li>\n\n\n\n<li><strong>Oscar Ramos-Soto<\/strong><br>Universidad de Guadalajara, Mexico<br>oscar.ramos9279(at)alumnos.udg.mx<\/li>\n\n\n\n<li><strong>Fernando Lezama<\/strong><br>GECAD &#8211; Polytechnic of Porto, Portugal<br>flz(at)isep.ipp.pt<\/li>\n\n\n\n<li><strong>Saul Zapotecas-Mart\u00ednez\u00a0<\/strong><br>Instituto Nacional de Astrof\u00edsica, \u00d3ptica y Electr\u00f3nica, Mexico\u00a0<br>szapotecas(at)@inaoep.mx<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Artificial intelligence (AI) systems are increasingly deployed in settings where predictive accuracy alone is insufficient. Their decisions must also be understandable, reliable, robust, and open to scrutiny. However, many explainability and trustworthiness problems involve competing objectives, complex search spaces, discrete or structured representations, and constraints that are difficult to address with conventional optimization methods. Evolutionary [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":36,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-1692","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/pages\/1692","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/comments?post=1692"}],"version-history":[{"count":2,"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/pages\/1692\/revisions"}],"predecessor-version":[{"id":1694,"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/pages\/1692\/revisions\/1694"}],"up":[{"embeddable":true,"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/pages\/36"}],"wp:attachment":[{"href":"https:\/\/www.evostar.org\/2027\/wp-json\/wp\/v2\/media?parent=1692"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}