This image showsKonstantin  Nikolaou

Konstantin Nikolaou

PhD Student
Institute for Computational Physics

Contact

Allmandring 3
70569 Stuttgart
Room: 1.077

  1. Nikolaou, K., Krippendorf, S., Tovey, S., & Holm, C. (2026). Beyond scaling curves: internal dynamics of neural networks through the NTK lens. Machine Learning: Science and Technology, 7, Article 2. https://doi.org/10.1088/2632-2153/ae4442
  2. Drotleff, J., Tovey, S., Hohenberger, P., Lohrmann, C., Hoßbach, J., Nikolaou, K., & Holm, C. (2026). Reinforcement Learning Enables Autonomous Microrobot Navigation and Intervention in Simulated Blood Capillaries. https://arxiv.org/abs/2606.26154
  3. Nikolaou, K., Scheunemann, J., Krippendorf, S., Tovey, S., & Holm, C. (2026). Spectral Reach: Understanding Neural Scaling as Progress into the Spectral Tail. https://arxiv.org/abs/2605.31244
  4. Tovey, S., Krippendorf, S., Spannowsky, M., Nikolaou, K., & Holm, C. (2025). Collective variables of neural networks: empirical time evolution and scaling laws. Machine Learning: Science and Technology, 6, Article 3. https://doi.org/10.1088/2632-2153/adee76
  5. Tovey, S., Lohrmann, C., Merkt, T., Zimmer, D., Nikolaou, K., Koppenhöfer, S., Bushmakina, A., Scheunemann, J., & Holm, C. (2025). SwarmRL: building the future of smart active systems. The European Physical Journal E, 48, Article 4. https://doi.org/10.1140/epje/s10189-025-00477-4
  6. Nikolaou, K., Krippendorf, S., Tovey, S., & Holm, C. (2025). Beyond Scaling Curves: Internal Dynamics of Neural Networks Through the NTK Lens. https://arxiv.org/abs/2507.05035
  7. Tovey, S., Lohrmann, C., Merkt, T., Zimmer, D., Nikolaou, K., Koppenhöfer, S., Bushmakina, A., Scheunemann, J., & Holm, C. (2024). SwarmRL: Building the Future of Smart Active Systems. https://doi.org/10.48550/arXiv.2404.16388
  8. Tovey, S., Krippendorf, S., Nikolaou, K., & Holm, C. (2023). Towards a phenomenological understanding of neural networks: data. Machine Learning: Science and Technology, 4, Article 3. https://doi.org/10.1088/2632-2153/acf099
To the top of the page