Interpretable Artificial Intelligence for Topological Photonics
We use sparse neural networks and KANs to obtain interpretable design principles governing band topology in photonic crystals, creating a dataset of photonic crystals with deterministic band topologies below the first gap.
Authors
- Thomas Christensen (ORCID: https://orcid.org/0000-0002-9131-2717)
- Sachin Vaidya (ORCID: https://orcid.org/0000-0002-3569-6568)
- Charlotte Loh (ORCID: https://orcid.org/0000-0003-4587-808X)
- Ziming Liu (ORCID: https://orcid.org/0000-0003-3581-7494)
- Marin Soljačić (ORCID: https://orcid.org/0000-0002-7184-5831)
- Ali Ghorashi (ORCID: https://orcid.org/0000-0002-5884-1756)
- Max Tegmark (ORCID: https://orcid.org/0000-0001-7670-7190)
Publication Details
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1364/cleo_fs.2025.ff138_5
- Primary Topic
- Photonic Crystals and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00