↓ Skip to main content
Chaithanya Naik Mude

Enhanced Neutral Atom Readout with ML

238 words·2 mins

Neutral-atom platforms read out qubits by imaging atom fluorescence — a noisy, camera-limited process. This project builds modular and scalable machine-learning models for neutral-atom qubit readout that leverage each site’s error profile, and an image-denoising front end that enables fast and accurate readout. Presented at the APS March Meeting 2025.

Key results
#

  • 70× reduction in model size compared to the state-of-the-art solution, at matched accuracy.
  • Matched-filter measurement of neutral-atom qubits (with the Saffman and Gauthier groups), published in Physical Review Applied.

Papers
#

Enabling Fast and Accurate Neutral Atom Readout through Image Denoising Chaithanya Naik Mude, Linipun Phuttitarn, Satvik Maurya, Kunal Sinha, Mark Saffman, Swamit Tannu. Preprint. [arXiv]

Efficient Measurement of Neutral-Atom Qubits with Matched Filters Robert M. Kent, Linipun Phuttitarn, Chaithanya N. Mude, Swamit Tannu, Mark Saffman, Gregory Lafyatis, Daniel J. Gauthier. Physical Review Applied. [arXiv]

Code
#

Code available upon request — public release coming soon.

BibTeX
#

@article{mude2025denoising,
  title   = {Enabling Fast and Accurate Neutral Atom Readout through Image Denoising},
  author  = {Mude, Chaithanya Naik and Phuttitarn, Linipun and Maurya, Satvik and Sinha, Kunal and Saffman, Mark and Tannu, Swamit},
  journal = {arXiv preprint arXiv:2504.08170},
  year    = {2025}
}

@article{kent2025matched,
  title   = {Efficient Measurement of Neutral-Atom Qubits with Matched Filters},
  author  = {Kent, Robert M. and Phuttitarn, Linipun and Mude, Chaithanya N. and Tannu, Swamit and Saffman, Mark and Lafyatis, Gregory and Gauthier, Daniel J.},
  journal = {Physical Review Applied},
  year    = {2025}
}