Machine-learning based high-bandwidth magnetic sensing
Machine learning tools were used to improve Nitrogen Vacancy (NV) magnetic sensing by tackling the trade-off between sensitivity and bandwidth in large dynamic range scenarios. The model, a Multilayer Perceptron, was trained on Electron Spin Resonance (ESR) data to predict resonance frequencies. Results show the ML approach can reduce required data points by at least a factor of three while maintaining error levels.