Fig. 1. A general quantum metrology device consists of quantum state preparation, parameter-encoding evolution and measurement. Here, the device is equipped with an ML agent. The agent associates to each measurement result a Bayesian distribution obtained from a neural network trained with calibration data. When acquiring repeated measurement with results , Bayesian distributions are multiplied, updating the prior knowledge about the unknown parameter . Finally, the agent chooses computes a phase for an adaptive feedback control of the device. Notice that and can be single- (as in the schematic here) or multivalued (as in the experiment of Cimini et al.17).