Published in ML Reproducibility Challenge 2022, 2023
Analysis of the Interpretability of Graph Neural Networks through the SubgraphX algorithm
Recommended citation: Mahlau, Y., Berg, L., & Kayser, L. (2023, July). [Re] On Explainability of Graph Neural Networks via Subgraph Explorations. In ML Reproducibility Challenge 2022.
Download Paper | Download Bibtex
Published in 41st International Conference on Machine Learning (ICML), 2024
Albatross is an extension of AlphaZero for simultaneous games
Recommended citation: Mahlau, Y., Schubert, F. & Rosenhahn, B. (2024, July). Mastering zero-shot interactions in cooperative and competitive simultaneous games. In Proceedings of 41st International Conference on Machine Learning (ICML).
Download Paper | Download Bibtex
Published in ACS Omega, 2025
A Memory-Efficient Gradient Computation Scheme using Time Reversibility of Maxwells equations
Recommended citation: Schubert, F., Mahlau, Y., Bethmann, K., Hartmann, F., Caspary, R., Munderloh, M., ... & Rosenhahn, B. (2025). Quantized inverse design for photonic integrated circuits. ACS omega, 10(5), 5080-5086
Download Paper | Download Bibtex
Published in SPIE, 2025
Presentation of the open-source framework FDTDX
Recommended citation: Mahlau, Y., Schubert, F., Bethmann, K., Caspary, R., Lesina, A. C., Munderloh, M., ... & Rosenhahn, B. (2025, March). A flexible framework for large-scale FDTD simulations: open-source inverse design for 3D nanostructures. In Photonic and Phononic Properties of Engineered Nanostructures XV (Vol. 13377, pp. 40-52). SPIE.
Download Paper | Download Bibtex
Published in Reinforcement Learning Journal, 2025
Inverse Design of 2D/3D Geometries using Multi-Agent Reinforcement Learning
Recommended citation: Mahlau, Y., Schier, M., Reinders, C., Schubert, F., Bügling, M., & Rosenhahn, B. (2025). Multi-Agent Reinforcement Learning for Inverse Design in Photonic Integrated Circuits. arXiv preprint arXiv:2506.18627.
Download Paper | Download Bibtex