SECAI School of Embedded Composite AI’s Post

Three research projects by Andrea Celli Alessandro Pigati, and Carlo Schwarz have each been awarded #ERC Starting #Grants, a competitive funding from the European Research Council (ERC) for early-career researchers. These projects contribute to Bocconi’s total of 64 ERC-funded #research initiatives, making it the leading European institution in the "Individuals, Markets and Organizations" disciplinary area, with 34 projects. In 2024, ERC #awarded 494 Starting Grants, amounting to 780M euros. Italian institutions received 41 grants, while Italian researchers (including those) secured 61, placing Italy second behind Germany in researcher nationality rankings. A total of 3,474 proposals had been submitted, with a success rate of 14.2%. #Bocconi, with 3 projects accepted out of 7 proposed, is therefore with 42% well above this European average Rector Francesco Billari notes that the diversity of ERC projects—from social sciences to mathematics and AI—highlights our increasingly quantitative approach to social sciences, incorporating machine learning and computational science to prepare future-proof students. The PLA-STEER project, led by Andrea Celli, focuses on scenarios where multiple machine learning algorithms interact on shared platforms. Typically, these platforms play a passive role, allowing decentralized learning. PLA-STEER aims to develop theoretical foundations for platforms that actively influence interactions, leading to more predictable outcomes while improving fairness and social welfare. The research will apply online learning and game theory to understand learning dynamics, tackle computational challenges, and create new algorithms to manage these enhanced platforms. Alessandro Pigati leads the MAGNETIC project, which focuses on solving complex mathematical problems involving "minimal submanifolds," special shapes and surfaces in higher dimensions. These shapes are key to understanding the link between small-scale geometry and large-scale structure. While significant progress has been made in simpler cases, much remains unknown in more complex, higher-dimensional scenarios. The project aims to study the behavior of these shapes, especially when they are unstable or don't minimize area, and explores their connection to physical phenomena where energy density concentrates around them. Carlo Rasmus Schwarz leads the CHAIN project, which investigates how social networks shape behavior across different groups: scientists, politicians, and citizens. The project consists of three studies. The first explores how social media, especially X, affects scientists' work by analyzing economists' online activity to assess how their networks influence information and research trends. The second one examines U.S. congressional staffers, analyzing how their backgrounds and networks influence policy decisions and voting behaviors. The third study focuses on India, analyzing how WhatsApp's policy changes impact beliefs and election outcomes.

PhD student in Computing Science with focus on serverless edge AI Project description Serverless edge AI is a transformative shift in computing architecture, merging serverless edge computing and artificial intelligence (AI) to harness and abstract the distributed cloud-edge resources. Unlike traditional serverless solutions, such as Function as a Service (FaaS) technology, which struggle with real-time adaptability, efficient function placement, and cost management, serverless edge AI allows AI models to be deployed closer to data sources, enhancing performance with lower latency, faster response times, scalability, and stronger privacy controls. This approach abstracts infrastructure management, enabling real-time AI processing directly where data is generated or consumed, meeting critical demands for reliability, privacy, and energy efficiency. This project aims to investigate and develop a serverless edge AI framework that addresses key challenges, such as enabling learning and inference at the edge for efficient resource orchestration, dynamically relocating workloads, supporting FaaS application mobility, and forecasting energy use and workload distribution across the cloud-edge continuum. By building an optimized framework with advanced algorithms (e.g., deep neural networks, sequence modeling, and multi-objective optimization), this project will enhance edge-based learning and decision-making, allowing for efficient orchestration of distributed resources. Key functionalities include intelligent serverless function execution, smart edge cluster provisioning, on-demand workload relocation, and device-specific model personalization. The framework and algorithms will be rigorously tested in real-time and simulation environments, demonstrating notable advantages over traditional centralized cloud models. Link to application: https://lnkd.in/eRETFiRW

🎉 ERASMUS Mundus Joint Masters "MATHS-DISC" : Application opened! 🚀 Fantastic news! The new website www.mathsdisc.eu was launched today and the application procedure for the first cohort starting in autumn 2025 was opened. 🌟 The MATHS-DISC consortium – comprising five renowned universities from across Europe, spanning Portugal to Ukraine (Università di Verona, Bergische Universität Wuppertal, Igor Sikorsky Polytechnic Institute Kyiv, Universidade de Lisboa, and Université Grenoble Alpes), alongside our esteemed partner university NTNU Trondheim – has successfully secured approval for an Erasmus Mundus Joint Master (EMJM) in Mathematical Modelling, Simulation, and Data Science. This groundbreaking program is supported by over 20 leading industry and research partners, including Siemens, Intel, Bosch, Bayer, d-fine, uRoboptics, Vodafone Ukraine, IDOM, ENGYS, Otto Bock, DLR, Terranova, NeuronsLab, Computer Solutions SA, Royal IHC, and more. Together, we aim to revolutionize interdisciplinary education and foster the PhD leaders of tomorrow! 💡 With a total budget of approximately 5 million Euros, MATHS-DISC is backed by the European Consortium for Mathematics in Industry (ECMI) and the European Network of Mathematics for Industry and Innovation (EU-MATHS-IN). Our mission is to combat geographical and scientific fragmentation by establishing robust European networks and sharing best practices in applied mathematics. 🔗 Now Online: mathsdisc.eu We are thrilled to announce that the MATHS-DISC website is now live at mathsdisc.eu! This will serve as the central hub for all program details, news, and updates. 📢 Applications Open for Autumn 2025 Intake We are officially opening the application procedure for the first cohort, starting in autumn 2025. If you're passionate about cutting-edge topics like Machine Learning, Digital Twins, Quantum Computing, and the traditional pillars of applied mathematics – Modeling, Stochastics, Numerical Simulation, Optimization – we invite you to apply and join this transformative journey. A special thanks to @José García (Piedrapapeltijera) for designing the new webpage. Great work! Stay tuned for more updates, and don’t forget to check out our website for detailed information about this unique EMJM program! 💻 Explore more at www.mathsdisc.eu. 📢 Let’s shape the future of applied mathematics together! #Mathematics #MATHSDISC #ErasmusMundus #JointMasters #DataScience #AI #MachineLearning #FutureLeaders #TransformativeEducation #EMJM #Europe #Ukraine

I had the pleasure of participating as a student in the panel discussion on shaping future policies for generative AI in Sri Lankan Higher Education. The event was organized by the General Research Committee (GRC) and Section E3 (Computer Science) of SLAAS, in collaboration with International Conference on Advanced Research in Computing,IEEE Sri Lanka Section , IEEE Computer Society Sri Lanka Chapter, SLAAS Sabaragamuwa Chapter, and Lanka Education And Research Network (LEARN) . It was an enlightening experience to engage in discussions on the future of AI in education. This insightful discussion provided valuable knowledge of; *opportunities and challenges of Gen AI *Policies *Future of Personalized Learning and Practices in Gen AI *Gen AI for transformative efficient learning *Approaches to Assessment *Best Practices for Reaserch Using Gen AI *Ethical Consideration *Guidelines for Gen AI use in education for student and concerns in using AI As a participant in this event, I would like to extend my heartfelt gratitude to the distinguished resource panel: Prof. Asoka Karunananda(Senior Professor, Faculty of Information Technology ,University of Moratuwa ) Dr. Asitha Bandaranayake(Senior Lecturer, University of Peradeniya,LEARN) Prof. Nuwan Kodagoda(Pro Vice Chancellor ,Faculty of Computing,SLIIT) Dr. Thushari Silva(Senior Lecturer,University Of Moratuwa,SLAAI) Dr. Menaka Ranasinghe(Seniors Lecturer,Electrical and Computer Engineering, Open University of sri lanka) Ms. Vishaka Nanayakkara(Senior Lecturer,University of Moratuwa,LIRNEasia ) Prof.Roshan Ragel (Professor,Head Department Of Computer Engineering ,University of Peradeniya,LEARN ) Dr. KPN. Jayasena(Postdoctoral Researcher,TU Dresden, Germany ) Your expertise, dedication, and insightful contributions were truly inspiring and instrumental in making this event a success. Thank you for sharing your knowledge and helping shape the future of AI in education #GenerativeAI #AIinEducation #HigherEducation #SriLanka #PanelDiscussion #AI #EducationPolicy #Innovation #FutureOfEducation #IEEE #SLAAS #Learning #StudentExperience #AcademicExcellence

I am proud to announce that I have successfully completed my MSc (Hons.) in Computer Science and Artificial Intelligence at Loughborough University acheiving a 2:1! 🎉 I am deeply grateful for the support and guidance from my professors, mentors, and peers who have been instrumental in my academic and personal growth. Here’s to the next chapter! 🚀 #MSc #ArtificialIntelligence #ComputerScience #LoughboroughUniversity #AI #MachineLearning #Graduation

The meeting between officials from The Johns Hopkins University and Union Education Minister Dharmendra Pradhan covered a range of topics, including faculty and student exchange programs, dual and combined degrees, and research collaborations in areas such as data science, AI, and emerging technologies. Read More 👉 https://lnkd.in/gVGw4pu5 #IndiaUSCollaboration #JohnsHopkinsUniversity #AcademicPartnerships #EducationInnovation #ResearchCollaboration #EmergingTechnologies #AIandDataScience #GlobalEducation #EPNews

🎓 Thrilled to announce my graduation! I am proud to share that I have officially obtained my Master’s degree in ´Information Processing and Complexity of Living Systems’ at Ecole Nationale d'Ingénieurs de Tunis in joint diploma with the Master’s program in Mathematics and Computer Science – Vision and Intelligent Systems at Université Paris Cité. For my master’s thesis, I focused on an explainable deep learning models for smoke detection in video sequences. This research leveraged a two-stream network combining spatial (frames) and temporal (optical flow) features to enhance accuracy and robustness. To ensure transparency and interpretability, I applied Grad-CAM and Shapley values to visualize and understand how the model made its predictions. This journey has been a challenging yet rewarding experience, and I’m deeply grateful to my professors, mentors, and peers for their guidance and support. Looking forward to applying these skills to impactful projects and contributing to innovations in AI! #Graduation #ExplainableAI #ComputerVision #DeepLearning #TwoStreamNetworks #GradCAM #SHAP #SmokeDetection

😍 "If your interest changes that is ok, please go ahead and change what you are doing" "AI is Intelligence Amplifiers" "Educability: A model of human cognition" "Coding is not programming" "Mathematics is like music" "Simplicity leads to small errors" "Discrete or continuous: two worlds of math" "Work on bridging two different areas" "A pragmatic view of the impact of AI" "The promise and perils of AI" 😍 ~ by Laureates. Was also great catching up with Prof. Yannis Ioannidis, President ACM. Pics: 🤩 With Prof. Pat Hanrahan (Turing Award, 2019), Prof. Les Valiant (Turing Award, 2010), Prof. David Patterson (Turing Award, 2017) & Prof. Yannis Ioannidis, President ACM, Association for Computing Machinery. 🤩 #ProfGiri #HLF24 #Heidelberg #Germany /c Heidelberg Laureate Forum Foundation, Heidelberg University

Don’t miss the opportunity: AI students who are starting their master’s studies at Technische Universität Dresden or Leipzig University or are already enrolled can apply for scholarships by the SECAI School of Embedded Composite AI until 30 June! SECAI is one of the three Konrad Zuse Schools of Excellence in Artificial Intelligence, which are financed with funds from the Bundesministerium für Bildung und Forschung. They focus on different aspects of AI research and provide excellent research-based education in Germany. With its scholarships, SECAI supports outstanding students whose studies are closely related to its research areas. These include, but are not limited to, the following MSc programmes: - Nanoelectronic Systems (TU Dresden) - Tracks “AI Engineering”, “Computational Life Science”, and “Logical Modelling” of Computational Modelling and Simulation (TU Dresden) - Data Science (Leipzig University) - Bioinformatics (Leipzig University) - Medical Informatics (Leipzig University) Recipients will receive funding for the remaining duration of their studies according to the model study plan (“Regelstudienzeit”). More information on SECAI and the scholarships can be found here: ➡️ https://lnkd.in/e5y_YbHv #DAAD #ChangeByExchange #ZuseSchoolsAI #AI #AIResearch #AITalents #scholarship

I'm glad to share my second PhD (Cum Laude), in the field of Computer Science and Artificial Intelligence, by the Faculty of Informatics UPV-EHU, and that is entitled ‘Técnicas bio-inspiradas para el diseño y aplicación de modelos cognitivos en la industria digital’ ('Bio-inspired techniques for the design and application of cognitive models to the digital industry'). Many thanks to my advisors, Basilio sIERRA and Diego Galar Pascual, and to everyone who has supported me during this exciting journey. The Abstract is as follows: 'Bio-inspired techniques have been a source of inspiration in the scientific community for ages to provide advanced solutions to solve complex problems. Moreover, most of the innovations and creative inventions that are proposed by humans along our history are based on the observation of natural phenomena and the innate abilities of the biological organisms to survive and evolve. The industrial sector is also adopting these technological advances, supported by the digitization of the productive systems and the available information to be exploited. The present dissertation shows an exhaustive research on the bio-inspired techniques that have been applied during recent years to the industrial processes. The obtained insights are mainly focused on the analysis of the current machine learning and artificial intelligence developments in real industrial environments. Once the state of the art of these bio-inspired methods is established, a set of analytical and decision support tools is proposed. The main objective is thus to illustrate how to develop cognitive models that are able to learn, think and adapt to complex and changing industrial scenarios. As a result, the productive systems are more robust, efficient and resilient.' #digital #research #machinelearning #artificialintelligence #math #mathematics #statistics #operationresearch #datascience #dataengineering #donostia #euskalherrikouniversitatea #facultadinformaticasansebastian #university #deeplearning