Keynotes

• Marine Le Morvan, INRIA - Resaerch Scientist
  Title: Learning with Missing Values: Theoretical Insights and Practical recommendations
  Abstract: Missing values are ubiquitous in many fields such as health, business or social sciences. To date, much of the literature on missing values has focused on imputation as well as inference with incomplete data. In contrast, supervised learning in the presence of missing values has received little attention. In this talk I will explain the challenges posed by missing values in regression and classification tasks. In practice, a common solution consists in imputing the missing values prior to learning. We will first examine the theoretical foundations of Impute-then-Regress approaches, and then clarify if and when investing in advanced imputation methods leads to significantly better predictions in practice.
  Biography: Marine Le Morvan is a research scientist at Inria. She is passionate about using AI to tackle complex scientific and healthcare problems, ensuring that machine learning models are both powerful and reliable. Her research lies at the intersection of statistical learning and trustworthy AI, with a focus on:
  • Tabular foundation models, which unlock new possibilities through large-scale pretraining.
  • Model auditing, to enhance the trustworthiness and reliability of machine learning systems.
  • Learning from incomplete data, a challenge pervasive in fields like healthcare and social sciences.

• Aurélien Bellet, INRIA - Senior Researcher
  Title: Privacy in Machine Learning
  Abstract:
  Biography: Aurélien Bellet is a senior researcher at Inria France. His current research focuses on the design of privacy-preserving machine learning algorithms in centralized and decentralized settings. Aurélien has served as area chair for top conferences such as ICML, NeurIPS and AISTATS. He co-organized several international workshops on machine learning and privacy. He also co-organizes FLOW, an online seminar on federated learning.

• Silvia Tulli, HABS - Senior AI Researcher
  Title: Multimodal Biometric Intelligence: From Academic Research to Industry Applications
  Abstract: What if AI could monitor pain levels in non-verbal patients or recognize emotions through brain activity and physiological signals? This talk explores multimodal biometric AI through an industry lens, focusing specifically on the deeptech sector. This talk emphasizes the symbiotic relationship between academia and industry. Traditional academic research provides foundational theories and controlled experimental validation. Deeptech companies translate these insights into real-world applications, confronting the harsh realities of noisy signals, missing modalities, and environmental interference. This talk will explore how we work on EEG signals and other biometric signals and build predictive and explainable models for different applications: security during driving, pain monitoring, brain-based authentication, and marketing research that captures biometric reactions beyond declarative responses. To conclude, this talk will outline specific open questions and discuss internship opportunities for motivated students interested in working on a challenging domain and seeing their research grow into a product.
  Biography: Silvia Tulli holds a PhD in AI and robotics, obtained through a Marie Curie scholarship, and completed a postdoc at the Institute of Intelligent Systems and Robotics at Sorbonne University. She serves as an EU expert for reviewing deeptech startups and has taught and conducted research in several European laboratories, including ENSTA - Institut Polytechnique de Paris, GAIPS at Instituto Superior Técnico in Portugal and CHILI at EPFL in Switzerland. During her interdisciplinary master’s degree in computer engineering and cognitive science at the University of Trento, she worked at Witted S.r.l. on underwater drones and later at CNRS in Pisa on assistive robotics. Her background has naturally translated into expertise in multimodal AI systems and sequential decision making—skills that are valuable for addressing complex signal processing challenges at HABS, where she currently collaborates as a senior AI researcher.

• Sihem Cherrared, Orange - Research Enginner
  Title: The Application of Agentic AI in Network Automation
  Abstract: The presentation explores the transformative potential of agentic AI in the realm of network automation. The platform leverages intelligent assistance through advanced intentionanalysis mechanisms, allowing clients to describe their needs in natural language. These need are then seemlessly translated into precise network configuration using Large Language Models (LLMs). The result is a powerful end-to-end automated process that enhances efficiency and reduces manual effort for networkmanagement. This innovative approach marks a significant step forward in deploying AI-driven solutions for network automation.
  Biography: Sihem Cherrared received a master’s degree in networks and distributed systems from the University of Science and Technology, Algeria, in 2015. She defended her PhD degree in 2020 at the University of Rennes 1, France, in INRIA and Orange Labs, focusing on the fault management of programmable multi-tenant networks. She held an R&D enginner position at Smile in Paris before joining Orange Innovation in 2022. She is currently working as research enginner and project leader on network automation. She is member of the IEEE ComSoc AI driven communications industry community. The topics of interests are the intent-based management and network automation.