This tutorial focuses on scientific experiments to enhance approaches, methodologies and models presented in scientific articles. We will study different typologies of publications to showcase the content of your research. Based on case studies we develop a methodology to structure the experimentation process.
We propose in this tutorial an experience on designing experiments for a specific scientific article, paper submission on easychair, reviewing articles by participating in a program committee (PC) dedicated to the event. (lecture and practical work)

This course will consist of two parts separated by a break: one on the Digital Services Act (DSA). This text, signed by the Council and the European Parliament on October 19, 2022, aims to reduce the distribution of illegal content and establish greater transparency between online platforms and their users, and a section on the laws adapting labor law and the information obligations of digital platforms.

In this tutorial we will cover the latest and most relevant approaches to extract faithful explanations from complex supervised machine learning models, perceived as black boxes. These include both explainable-by-design models, as well as post-hoc explainability layers. We will discuss the different paradigms to open the black box, the challenges of doing so, as well as the methodological tools to evaluate the quality of the explanations both from a functional and user perspective. We will have a quick practical session where the participants will be able to compute explanations for the inner-workings of different black-box models. (lecture and practical work)

Recommendation systems play a pivotal role in guiding users through an abundance of content, products, and services available online. However, traditional algorithms often suffer from the popularity bias and fail to address the diverse needs and preferences of users. In this hands-on tutorial, we delve into the realm of diversity algorithms tailored for recommendation systems. Beginning with an overview of recommendation systems and the significance of diversity, participants will gain insights into the challenges and the benefits of incorporating diverse recommendations. We will then explore various diversity metrics and algorithms. Through a combination of theoretical discussions and practical exercises, attendees will have the opportunity to implement and experiment with diversity algorithms using movie datasets.

(lecture) The first part will introduce the main theoretical frameworks: (1) the Earth system and the framework of planetary limits, (2) Kate Raworth's Donut model and (3) methods for designing digital tools in resource-constrained contexts. In the second part, participants will be able to put these elements into practice in collaborative workshops for the (re)design of digital tools linked to large quantities of data.

In the past few years, there has been much work on incorporating fairness requirements into the design of algorithmic rankers, with contributions from the data management, algorithms, information retrieval, and recommender systems communities. In this tutorial, I will give a systematic overview of this work, offering a broad perspective that connects formalizations and algorithmic approaches across subfields. During the first part of the tutorial, I will present a classification framework for fairness-enhancing interventions, along which I will then relate the technical methods. This framework helps unify the presentation of mitigation objectives and of algorithmic techniques to help meet those objectives or identify trade-offs. Next, I will discuss fairness in score-based ranking and in supervised learning-to-rank. I will conclude with recommendations for practitioners, to help select a fair ranking method based on the requirements of their specific application domain.

In this talk, I invite you to accompany me on a transformative journey, one that began in the world of code and led me to the realm of remote leadership of a diverse and distributed team of software engineers, transitioning from a role focused on technical expertise to one centered on managing people, projects, and strategies.
My story is a testament to the idea that a career in technology can evolve and expand beyond coding, presenting opportunities for personal and professional growth into leadership roles.
Through my narrative, you will gain insights into the unique challenges faced by engineers stepping into leadership positions. We will discuss the adjustments needed in mindset and skillset, and we will share practical advice for those considering or navigating a similar journey.
Whether you are a seasoned developer looking to expand your horizons, an aspiring manager seeking guidance, or a current leader eager to empathize with the experiences of your team members, join me as we unravel the story of my transition from software engineer to manager, and discover the keys to successful leadership in the ever-evolving tech industry.

Aviation, like many other technical fields, is lacking in diversity. Join former military and commercial pilot Jessica Brightman to discuss the male-dominated aviation and drone industries. She'll discuss her current efforts to improve inclusivity in the emerging drone industry and aviation as a whole, as well as the history of women in aviation, beginning with the first certificated female pilot in 1910, the first female airline pilot in 1969, and the current statistics of women in aviation.

After more than a century of powered flight, the time is right to push to increase diversity, equity, inclusion, and accessibility in aviation, and the drone industry is leading the way. Jessica will share her thoughts about what we all can do to improve aviation and other technical fields for everyone.