# ML Fine Tuning of Quantum Dots

In this video we cover the concept of fine-tuning of semiconductor quantum dot devices. We discuss the wealth of fine-tuning task and highlight Hamiltonian learning as an example we will follow for the rest of this module.

# Main takeaways

- Fine-tuning covers very diverse set of tasks and is highly dependent of specific experimental setup and measurement
- Useful abstraction for many fine-tuning tasks is that of mapping measurement data on a parameter in some underlying abstract model we are aiming to experimentally reproduce

# Further thinking

Which of the following is NOT an example of fine-tuning:

a. Hamiltonian learning

b. Gate tuning

c. Clustering

d. Tunnelling rates adjustment

# Further reading

Suppressing qubit dephasing using real-time Hamiltonian estimation: https://www.nature.com/articles/ncomms6156

Hamiltonian estimation using the Bayesian inference method of the Overhauser field.

Spectrum of the Nuclear Environment for GaAs Spin Qubits: https://arxiv.org/abs/1701.01855

This experiment recovers the Overhauser field from the qubit precession. Main figures in the notebooks are taken from this paper.

Supplementary information for ”Spectrum of the Nuclear Environment for GaAs Spin Qubits”: https://journals.aps.org/prl/supplemental/10.1103/PhysRevLett.118.177702/Overhauser_supplement.pdf

Measurement of Temporal Correlations of the Overhauser Field in a Double Quantum Dot: https://arxiv.org/abs/0712.4033

A classical model of Overhauser field fluctuations due to nuclear spin diffusion is used to fit the experimental data in Fig. 2(c) in the notebook.

Characterizing non-Markovian Quantum Processes by Fast Bayesian Tomography: https://arxiv.org/abs/2307.12452

Experimental implementation of future prediction for Overhauser Field Gradient.