Abstract: Graph neural networks (GNNs) provide a natural framework for modeling systems of interacting agents, where dynamics and decisions are determined by local interactions. Yet, training GNNs on large-scale networks remains a challenge due to the lack of Euclidean structure and the computational burden of graph learning. In this talk, I will discuss learning by transference, a framework that enables scalable learning and control across graph families with shared structural properties. The key idea is that networks sampled from a common generative model, such as a graphon or manifold, exhibit similar dynamics, allowing GNNs trained on smaller graphs to generalize to larger ones.
I will show how this transferability principle can be leveraged to learn decentralized control policies for multi-agent systems. In particular, I will present empirical results on flocking, where GNNs learn coordination policies that generalize across different network sizes and topologies. These results illustrate how graph-based imitation learning can achieve efficient, scalable, and robust control.
Bio: Luana Ruiz received the Ph.D. degree in electrical engineering from the University of Pennsylvania in 2022, and the M.Eng. and B.Eng. double degree in electrical engineering from the École Supérieure d’Electricité and the University of São Paulo in 2017. She is an Assistant Professor with the Department of Applied Mathematics and Statistics and the MINDS and DSAI Institutes at Johns Hopkins University, as well as the Electrical and Computer Engineering and Computer Science departments (by courtesy). Luana’s work focuses on large-scale graph information processing and graph neural network architectures. She was awarded an Eiffel Excellence scholarship from the French Ministry for Europe and Foreign Affairs between 2013 and 2015; nominated an iREDEFINE fellow in 2019, a MIT EECS Rising Star in 2021, a Simons Research Fellow in 2022, and a METEOR fellow in 2023; and received best student paper awards at the 27th and 29th European Signal Processing Conferences. Luana is currently a member of the Machine Learning for Signal Processing Technical Committee of the IEEE Signal Processing Society.
Host: Abhishek Cauligi