Nikolaos Nakis

Postdoctoral Researcher @ Yale University

About Me

Profile picture of Nikolaos Nakis

I hold a PhD from the Technical University of Denmark (DTU) in the Section of Cognitive Systems at DTU Compute. Following my doctoral studies, I was a postdoctoral researcher at École Polytechnique in the Polytechnic Institute of Paris, where I worked in the Data Science and Mining Lab. Currently, I am a postdoctoral researcher at the Yale Institute for Network Science in the Human Nature Lab. My research focuses on machine learning for graphs, particularly graph representation learning and statistical learning. I apply these methods to areas such as social network analysis, network polarization, and developing self-explainable models to enhance interpretability and transparency in AI systems.

Publications

  • Continuous-time Graph Representation with Sequential Survival Process (2024)
    Abdulkadir Çelikkanat, Nikolaos Nakis, Morten Mørup
    Proceedings of the AAAI Conference on Artificial Intelligence, 2024
    View Paper
  • Time to Cite: Modeling Citation Networks using the Dynamic Impact Single-Event Embedding Model (2024)
    Nikolaos Nakis, Abdulkadir Çelikkanat, Louis Boucherie, Sune Lehmann, Morten Mørup
    Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR
    View Paper
  • A Hierarchical Block Distance Model for Ultra Low-Dimensional Graph Representations (2023)
    Nikolaos Nakis, Abdulkadir Çelikkanat, Sune Lehmann, Morten Mørup
    IEEE Transactions on Knowledge and Data Engineering (TKDE)
    View Paper
  • Characterizing Polarization in Social Networks using the Signed Relational Latent Distance Model (2023)
    Nikolaos Nakis, Abdulkadir Çelikkanat, Louis Boucherie, Morten Mørup
    Proceedings of The 26th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR
    View Paper
  • A Hybrid Membership Latent Distance Model for Unsigned and Signed Integer Weighted Networks (2023)
    Nikolaos Nakis, Abdulkadir Çelikkanat, Morten Mørup
    Advances in Complex Systems (ACS)
    View Paper
  • HM-LDM: A Hybrid-Membership Latent Distance Model (2022)
    Nikolaos Nakis, Abdulkadir Çelikkanat, Morten Mørup
    Complex Networks and Their Applications XI
    View Paper
  • Piecewise-Velocity Model for Learning Continuous-time Dynamic Node Representations (2022)
    Abdulkadir Çelikkanat, Nikolaos Nakis, Morten Mørup
    Proceedings of the First Learning on Graphs Conference, PMLR
    View Paper
  • Machine Learning for Static and Single-Event Dynamic Complex Network Analysis (2024)
    Nikolaos Nakis
    PhD Thesis, 2024
    View Paper

Preprints

  • Signed Graph Autoencoder for Explainable and Polarization-Aware Network Embeddings (2024)
    Nikolaos Nakis, Chrysoula Kosma, Giannis Nikolentzos, Iakovos Evdaimon, Michail Chatzianastasis, Michalis Vazirgiannis
    Preprint, 2024
    View Paper
  • Capturing Interactions in Complex Protein-Protein Networks: Introducing the Signed Two-Space Proximity Model (2024)
    Nikolaos Nakis, Chrysoula Kosma, Anastasia Brativnyk, Iakovos Evdaimon, Michail Chatzianastasis, Michalis Vazirgiannis
    Preprint, 2024

Awards

  • Best Paper Award
    "Continuous-time Graph Representation with Sequential Survival Process"
    Temporal Graph Learning Workshop @ NeurIPS 2023

Talks

  • "Characterizing Polarization in Social Networks using Archetypal Analysis"
    July 2023
    9'th International Conference on Computational Social Science, IC2S2
  • "The Hybrid-Membership Latent Distance Model"
    November 2022
    11'th International Conference on Complex Networks and their Applications, CNA

News

  • Paper accepted in AAAI! - 2024
  • Paper accepted in AISTATS! - 2024
  • Paper accepted in TKDE! - 2023
  • Paper accepted in AISTATS! - 2023