Improved Tomographic Estimates by Specialized Neural Networks

Characterization of quantum objects, being them states, processes, or measurements, complemented by previous knowledge about is a valuable approach, especially as it leads to routine procedures for real-life components. To this end, Machine Learning algorithms have demonstrated successfully operate in presence noise, estimating specific physical parameters. Here we show that neural network (NN) can improve the tomographic estimate parameters including convolutional stage. We applied our technique process tomography characterization several channels. demonstrate stable and reliable operation achievable training only with simulated data. The obtained results viability approach an effective tool based on completely new paradigm employment NNs operating classical data produced systems.