Amauri H. Souza
Scholar

Amauri H. Souza

Google Scholar ID: lP0LBI4AAAAJ
Federal Institute of Ceará (IFCE)
machine learningdeep learninggraph neural networks
Citations & Impact
All-time
Citations
5,382
 
H-index
14
 
i10-index
14
 
Publications
20
 
Co-authors
23
list available
Resume (English only)
Academic Achievements
  • Multiple papers accepted at NeurIPS'25, ICML'25, NeurIPS'23, ICML'24. Topics include persistent homology on graphs, topological positional encodings, generalization theory of equivariant GNNs, and more.
Research Experience
  • Professor at IFCE; Former postdoc at Cornell University and Aalto University.
Education
  • Postdoc at Cornell University (USA), under Kilian Weinberger; Postdoc at Aalto University (Finland), under Vikas Garg and Samuel Kaski.
Background
  • Professor at the Department of Computer Science of the Federal Institute of Ceara (IFCE). Research interests include advancing machine learning to deal with structured data (e.g., graphs and sequences) in several application domains such as recommender systems, computational social science, computational finance, and drug discovery. Recent works cover topological & geometric deep learning, explainable AI, temporal graph learning, simulation-based inference, generative modeling (GFlowNets), and graph neural networks.
Miscellany
  • Three PhD positions open, application deadline: November 7, 2025; Two M.Sc. positions open, topics include topological deep learning and temporal graph learning.