Graph Machine Learning Group
Graph machine learning • Non-stationary environments • Spatiotemporal data • Reinforcement learning • Dynamical systems
About GMLG
We are a research team part of the Swiss AI Lab (IDSIA) at Università della Svizzera italiana, Lugano, Switzerland. The group is led by Prof. Cesare Alippi.
The group was previously under the ALaRI institute.
Meet our teamNews
-
May 2026
Three papers accepted at ICML 2026: What Matters in Deep Learning for Time Series Forecasting? (Moretti et al.), Beyond Softmax: A Natural Parameterization for Categorical Random Variables (Manenti and Alippi) and SWING: Unlocking Implicit Graph Representations for Graph Random Features (Manenti et al). -
Feb 2026
Two new preprints out! What Matters in Deep Learning for Time Series Forecasting? (Moretti et al.) and SWING: Unlocking Implicit Graph Representations for Graph Random Features (Manenti et al). Have a look! -
Oct 2025
Two new preprints about our latest research! Beyond Softmax: A Natural Parameterization for Categorical Random Variables (Manenti and Alippi) and Hierarchical Message-Passing Policies for Multi-Agent Reinforcement Learning (Marzi et al). Check them out! -
Oct 2025
We have released two new preprints about Online Continual Graph Learning (OCGL, Donghi et al) and The Unreasonable Effectiveness of Randomized Representations in OCGL (Donghi et al). Take a look! -
Sep 2025
Our papers Over-squashing in Spatiotemporal Graph Neural Networks (Marisca et al.) and Equilibrium Policy Generalization: A Reinforcement Learning Framework for Cross-Graph Zero-Shot Generalization in Pursuit-Evasion Games (Lu et al.) have been accepted at NeurIPS 2025!
Research
Our research focuses on graph machine learning, non-stationary environments, dynamical systems, and reinforcement learning.
We also apply machine learning in many diverse fields, including neuroscience, power grids, chemistry, and agriculture, among many others.
Open Source
Our group is active in open source software development and we maintain several Python libraries based on our research. Check out also the group GitHub page for code related to our papers.
Software
