Publications
Frobenius Normalization for Quantum State Reconstruction — Under review at IEEE qCCL 2026
A quantum state describes the probabilities of a quantum system. The problem: systems decohere over time, losing information to the environment. I use neural networks to reconstruct full states from incomplete measurements. The challenge is that mixed states become smaller, requiring the model to learn both rotation and stretching. My solution uses Frobenius normalization to artificially scale inputs, enabling the model to handle larger systems.
Research Interests
- Distributed training
- Model stability
- Interpretability
- Quantum machine learning