Seminarios Curso 2025-2026

Título: Evolving Fuzzy Systems: Concepts and Applications to Drift Problems

Ponente: Hans Peter de Koning, NOVA IMS — Universidade Nova de Lisboa

Sesiones:

• 2 y 3 de Febrero de 2026, de 9h a 14h

Lugar: Aula 2.0.C03

Resumen:

Evolving Fuzzy Systems (EFS) integrate concepts from Artificial Neural Networks and Fuzzy Systems, combining their strengths to address complex and dynamic problems. Drift problems are associated with datasets whose underlying behavior changes over time, requiring adaptive modeling approaches. In this context, EFS provide suitable solutions for identifying and responding to such changes. This seminar will present the fundamental concepts of Evolving Fuzzy Systems and concept drift, supported by practical demonstrations using Python libraries, with the aim of offering students both theoretical understanding and hands-on experience.
 

Breve biografía:

Paulo Vitor de Campos Souza is an Invited Assistant Professor at NOVA IMS — Universidade Nova de Lisboa. He holds a Ph.D. (Technical Sciences, Computer Science emphasis) from JKU Linz. His research focuses on neuro-fuzzy models, evolving/online learning, and interpretable AI. He previously served as a Research Engineer at Fondazione Bruno Kessler (FBK), concluding his tenure there in August 2025.

Título: Software Reverse Engineering Crash Course: From Zero to F

Ponente: Simone Aonzo, EURECOM

Sesiones:

• 4 y 5 de marzo 2026, de 9 a 14h

Lugar: Aula 2.0.C03

Resumen:

This seminar offers a concise introduction to software reverse engineering, with an emphasis on static analysis of Windows executables.

Starting from first principles, participants learn how executables are constructed and how they interact with the operating system.

The seminar covers executable structure (the PE format), core computer architecture concepts, x86/x64 assembly fundamentals, and hands-on static analysis using the open-source reverse engineering framework Ghidra.

It also introduces techniques commonly used by malware, such as obfuscation, packing, and basic evasion strategies.

Special attention is given to state-of-the-art methods for assisting reverse engineering with Large Language Models.

By the end of the seminar, participants will be able to methodically analyze unknown Windows binaries, understand their structure and low-level behavior, identify common malware techniques, and leverage modern approaches to support and accelerate the reverse engineering process.
 

Breve biografía:

Simone Aonzo is an Assistant Professor at EURECOM (France), where he leads research in cybersecurity, focusing on Android and mobile system security, malware analysis, applied machine learning in security, and human factors in threat analysis.

He has published extensively in top venues, contributing impactful work on subjects such as malware ecosystem dynamics, the interplay between human and automated processes, phishing attacks, and limitations of machine-learning detectors.

Simone also serves the research community as a program committee member and reviewer for leading security conferences and journals.

In addition to his academic career, he brings practical experience as an Android penetration tester and malware analyst.

He continues to bridge research and practice as co-founder of ThreatNemesis, a startup developing advanced Android analysis sandbox technologies rooted in his research.

Título: Enterprise AI: Software Engineering Challenges in Deploying AI Technologies at Scale

Ponente: Christoph Bussler, Mistral AI

Sesiones:

• 14 y 15 de Abril de 2026, de 9h a 13h
• 16 de Abril 2026, de 9h a 11h

Lugar: Aula 2.0.C03

Resumen:

Since the launch of ChatGPT in November 2022, AI has rapidly moved from research labs into everyday applications—transforming search, communication, e-commerce, healthcare, and beyond. However, while consumer AI tools have seen widespread adoption, enterprises face unique challenges. For large organizations, the focus is not just on deploying AI, but on ensuring reliability, scalability, and measurable value in production environments.

This seminar defines what makes AI “enterprise-grade” and explores the technical areas where AI is making the biggest impact for businesses. We’ll dive into the engineering hurdles of productizing AI: from model selection and fine-tuning, to integration with legacy systems, to ensuring robustness, security, and compliance. Deploying AI in a commercial context requires more than state-of-the-art algorithms—it demands rigorous software engineering, DevOps practices, and a deep understanding of real-world constraints.
 

Breve biografía:

Christoph Bussler is an AI Solutions Architect at Mistral AI, where he specializes in deploying cutting-edge AI technologies for large enterprise clients. Passionate about bridging the gap between scientific research and real-world applications, Christoph thrives at the intersection of innovation and practical impact—transforming advanced AI into scalable software solutions that drive customer success.

Throughout his career, Christoph has made significant contributions to a range of transformative technologies, including database and integration systems in public clouds (Google Cloud), Semantic Web standards, Enterprise Application Integration (EAI), Business-to-Business Integration (B2B), and Workflow Management. Beyond implementation, he has authored several influential books and publications on these topics, sharing his expertise with the broader academic and software technology community.

Explore his professional work and publications at http://real-programmer.com.