Efficient Recursive Deep Graph Networks and Issues Analyses

Speaker: Prof. Alessio Micheli, University of Pisa

Abstract: Efficiently dealing with complex data structures such as sequences and graphs is a fascinating and ever-relevant challenge, with an impact on the applicability and environmental sustainability of deep neural networks. In relation to these issues, after a short introduction and recap of Deep Graph Networks, this talk will outline a research path based on efficient reservoir computing methods. We will pay particular attention to advancements in the analysis of models properties (stability, efficiency, expressiveness), including the use of explainability techniques (XAI) on graphs.

Biography: Alessio Micheli is Full Professor at the Department of Computer Science of the University of Pisa, where he is the head and scientific coordinator of the Computational Intelligence & Machine Learning Group (CIML), part of the CAIRNE.eu Research Network. His research interests include machine learning, neural networks, deep learning, learning in structured domains (sequence, tree, and graph data), recurrent and recursive neural networks, reservoir computing, and probabilistic and kernel-based learning for non-vectorial data, with a particular focus and pioneering works on efficient neural networks for learning from graphs. The applications area includes Health and Bio/ChemInformatics. In these research areas, he has authored (at 2025) over 250 articles. Prof. Micheli is the national coordinator of the "Italian Working group on Machine Learning and Data Mining" of the Italian Association for Artificial Intelligence and he has been co-founder/chair of the IEEE CIS Task Force on Reservoir Computing. He is the elected president of the European Neural Network Society – ENNS. He serves as an Associate Editor for Neural Networks and IEEE Transactions on Neural Networks and Learning Systems.

Host: Prof. Cesare Alippi