Carlo Abate

I am a Postdoctoral Researcher at the Department of Mathematics and Statistics at the Arctic University of Norway (UiT), where I work with Prof. Filippo Maria Bianchi on basic research on graph neural network methods and applications, with focus on graph pooling techniques and graph generative models.

I completed my Ph.D. in Data Science and Computation at the University of Bologna in collaboration with the Italian Institute of Technology (IIT) in Genoa. My research focused on developing innovative computational methods for drug discovery, particularly working on latent variable generative models for molecular graphs.

I am currently open to supervising Master’s and Ph.D. students, especially those who are interested in spending part of their Ph.D. period at UiT as visiting students. If you are interested in collaborating or have any inquiries, please feel free to get in touch!.

latest news

May 01, 2025	Starting as a Postdoctoral Researcher at the Department of Mathematics and Statistics, UiT.
Mar 26, 2025	Successfully defended my Ph.D. thesis in Data Science and Computation at the University of Bologna.
Jan 22, 2025	Our paper “MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks” has been accepted at ICLR 2025! 🎉

	show more

selected publications

ICLR
MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks

Carlo Abate, and Filippo Maria Bianchi

In The Thirteenth International Conference on Learning Representations, 2025

Abs Bib HTML Code

We propose a novel approach to compute the MAXCUT in attributed graphs, i.e. graphs with features associated with nodes and edges, by exploiting heterophilic message passing to assign connected nodes to different partitions. The approach is fully differentiable, making it possible to find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement MaxCutPool, a hierarchical graph pooling layer for graph neural networks. The layer is sparse, differentiable, and particularly suitable for downstream tasks on heterophilic graphs. Our key contributions include: 1) A novel MAXCUT computation method for attributed graphs, 2) A new hierarchical pooling layer especially effective for heterophilic graphs, 3) A general scheme for node-to-supernode assignment, 4) The introduction of the first heterophilic dataset for graph classification.
@inproceedings{abate2025maxcutpool, title = {MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks}, author = {Abate, Carlo and Bianchi, Filippo Maria}, booktitle = {The Thirteenth International Conference on Learning Representations}, year = {2025}, url = {https://openreview.net/forum?id=xlbXRJ2XCP}, }

latest news

selected publications

connect