Abstract: TBD. Talk on test-time training and reinforcement learning by Jonas Hübotter (ETH Zürich). Approximately 45 minutes.

Senior Machine Learning Engineer | Community lead | Board member | Machine Learning, AI solutions, Inclusive & Ethical Digital Services

Email campaign success through obtaining high engagement numbers has become harder for platforms with diverse user bases who operate at large scale. Traditional standardized email methods to deliver marketing content fall short because there are too many messages combined with brief viewer attention. Joint usage of personalized subject lines with optimized send timing functions as primary methods to gain user engagement thus leading to better open rates and substantive user interactions. The study establishes a complete data-based system which employs behavioral analytics, machine learning models together with A/B testing for dynamic personalization of subject lines and customized send schedules on an individual basis. The proposed system does that by utilizing historical user data, engagement patterns and contextual signals to send highly targeted, timely email contents to dramatically improve user engagement, increases the click through rates and conversion rate from email. Instead of having to sit down and think of a project strategy, the way this approach works is that personalization can be automated at scale as well as through continuous learning and optimization using real-time feedback loops and adaptive algorithms to provide a scalable solution for today’s modern email marketing challenges.

Discover how the Bundesliga leveraged AWS to translate and localize media content, allowing fans around the world to enjoy it in their native language. We will present the modular workflow designed to handle diverse content types, the different steps of the localization process, and the system's automatic learning capabilities.

The presentation introduces Smart Alerting, a system that uses machine learning and statistics to automatically detect anomalies in sales and performance data during platform rollouts. It helps identify incidents faster, reduce manual monitoring, and support data-driven decision-making.

Atelier de 2 heures sur l'extraction de données des sites web avec Python, automatisation de la collecte et de l'analyse, avec Jupyter notebooks.

To solve problems in science, engineering, and business, computers were first programmed with the explicit instructions to solve those problems. Now, AI has shown that it is more powerful to first train computers to learn and then give them the data needed to solve a problem. I will review the successes and limitations of machine learning methods being used to train: 1) quantum computers with only Dirac operator gates, 2) hybrid classical-quantum computers with variational quantum circuits, 3) quantum Hopfield computers using equilibrium propagation, a quantum replacement for back propagation, and 4) quantum computer annealers.

AI is rapidly transforming cybersecurity - used by both attackers and defenders to outsmart one another in real time. Ethical hackers now face a new frontier where AI-driven threats demand AI-powered defenses. A 60-minute live session exploring how AI is reshaping ethical hacking, evolving tactics of cybercriminals, and the strategies you need to safeguard organizations in this new era. What you’ll learn:\n1. How AI is being weaponized in modern cyberattacks.\n2. Defensive use cases of AI in ethical hacking.\n3. Real-world examples of AI vs. AI in action.\n4. Skills and tools every ethical hacker should master.

Abstract: There is great interest in scaling the number of tokens that LLMs can efficiently and effectively ingest, a problem that is notoriously difficult. Training LLMs on a smaller context and hoping that they generalize well to much longer contexts has largely proven to be ineffective. In this talk, I will go over our work that aims to understand the failure points in modern LLM architectures. In particular, I will discuss dispersion in the softmax layers, generalization issues related to positional encodings, and smoothing effects that occur in the representations. Understanding these issues has proven to be fruitful, with related ideas now already being part of frontier models such as LLaMa 4. The talk is intended to be broadly accessible, but a basic understanding of the Transformer architectures used in modern LLMs will be helpful.

Integrating machine learning with DevOps practices is essential for organizations to stay competitive. This hands-on workshop will introduce you to JFrog ML and its capabilities, empowering data scientists and DevOps teams to seamlessly manage the end-to-end machine learning lifecycle. Learn to securely build, deploy, and maintain machine learning models with JFrog’s powerful platform, while enhancing collaboration between data scientists and DevOps teams.

Join Sheamus for an in-depth webinar on the exciting intersection of artificial intelligence and robotics. This session will provide a foundational understanding of how AI is revolutionizing the field of robotics, moving beyond traditional, pre-programmed systems to create intelligent, autonomous machines. Sheamus will explore the core concepts of AI that are most relevant to robotics, including machine learning, computer vision, and natural language processing. The webinar will cover practical applications and case studies, from self-navigating drones to collaborative industrial robots. Attendees will gain insight into the challenges and opportunities in this rapidly evolving field, and learn about the key technologies and skills needed to design and build the next generation of intelligent robots. Whether you are a student, an engineer, or simply curious about the future of automation, this session will provide a comprehensive and accessible introduction.

An in-depth webinar on how AI is transforming robotics, moving beyond pre-programmed systems to autonomous machines. This session covers core AI concepts—machine learning, computer vision, and natural language processing—with practical applications and case studies from self-navigating drones to collaborative industrial robots.

Join Sheamus for an in-depth webinar on the exciting intersection of artificial intelligence and robotics. This session will provide a foundational understanding of how AI is revolutionizing the field of robotics, moving beyond traditional, pre-programmed systems to create intelligent, autonomous machines. Sheamus will explore the core concepts of AI that are most relevant to robotics, including machine learning, computer vision, and natural language processing. The webinar will cover practical applications and case studies, from self-navigating drones to collaborative industrial robots. Attendees will gain insight into the challenges and opportunities in this rapidly evolving field, and learn about the key technologies and skills needed to design and build the next generation of intelligent robots. Whether you are a student, an engineer, or simply curious about the future of automation, this session will provide a comprehensive and accessible introduction.

Join Sheamus McGovern for an in-depth webinar on the intersection of artificial intelligence and robotics. This session introduces foundational AI concepts relevant to robotics (machine learning, computer vision, and natural language processing) and explores practical applications and case studies, including self-navigating drones and collaborative industrial robots. Attendees will gain insights into challenges, opportunities, and the key technologies and skills needed to design and build the next generation of intelligent robots.

This project develops an enterprise-grade AI platform that automates the extraction of ESG data, regulatory compliance checks, and peer benchmarking for companies. Utilizing NLP and machine learning, the system converts unstructured sustainability reports into standardized metrics, facilitating real-time compliance monitoring and competitive intelligence across various industries. Business Impact: Targets the rapidly growing ESG software market, serving investment firms, consulting companies, and institutional investors requiring automated analysis for portfolio decisions and regulatory compliance.

Estimating weight of cars from their dimensions

A conversation with Dashel Ruiz Perez about how data shapes semiconductor manufacturing and engineering, lessons from teaching programming and ML at scale, first-hand experience with ML Zoomcamp, moving from dashboards to ML-powered applications, and advice for professionals transitioning into data and AI later in their careers.

Continuation of the Maths and Coding – Linear Algebra Session.

We will have a practical Maths and Coding – Linear Algebra Session covering fundamental linear algebra concepts essential in machine learning, data science, and numerical computing. The session will be facilitated by Jessica González, a mathematician and data analyst with a strong background in education and applied mathematics. Jessica is currently working as a Data Analyst at Radius Fuel Cards, where she builds dashboards, analyzes fuel consumption data, and improves pricing estimators. She holds a Master’s in Mathematics from Freie Universität Berlin and has previously worked as an IB Mathematics Instructor and University Assistant.

Metric analysis is hard. For engineers and for machines. I'll show you how to make it easy. In this talk, I'll share simple steps to boost your Grafana dashboard's UX and ML capabilities. Want to automate your Grafana analysis? Join my session to find out how.