Optical sensors based on quantum light need to work even in the presence of imperfections. Here we discuss how to address the presence of noise by measuring multiple parameters at once: those we seek to monitor and those linked to the imperfections. Our method is applied to the investigation of the optical activity of sugary solutions limited by reduced visibility. These studies introduce multiparameter estimation as a viable approach to allow for robust operation of quantum sensors as they are developed from fundamental research to technology.

Introducing quantum sensors as a solution to real world problems demands reliability and controllability outside of laboratory conditions. Producers and operators ought to be assumed to have limited resources readily available for calibration, and yet, they should be able to trust the devices. Neural networks are almost ubiquitous for similar tasks for classical sensors: here we show the applications of this technique to calibrating a quantum photonic sensor. This is based on a set of training data, collected only relying on the available probe states, hence reducing overhead. We found that covering finely the parameter space is key to achieving uncertainties close to their ultimate level. This technique has the potential to become the standard approach to calibrate quantum sensors.

Recent developments have led to the possibility of embedding machine learning tools into experimental platforms to address key problems, including the characterization of the properties of quantum states. Leveraging on this, we implement a quantum extreme learning machine in a photonic platform to achieve resource-efficient and accurate characterization of the polarization state of a photon. The underlying reservoir dynamics through which such input state evolves is implemented using the coined quantum walk of high-dimensional photonic orbital angular momentum and performing projective measurements over a fixed basis. We demonstrate how the reconstruction of an unknown polarization state does not need a careful characterization of the measurement apparatus and is robust to experimental imperfections, thus representing a promising route for resource-economic state characterization.

Calibration of sensors is a fundamental step in validating their operation. This can be a demanding task, as it relies on acquiring detailed modeling of the device, which can be aggravated by its possible dependence upon multiple parameters. Machine learning provides a handy solution to this issue, operating a mapping between the parameters and the device response, without needing additional specific information on its functioning. Here, we demonstrate the application of a neural-network-based algorithm for the calibration of integrated photonic devices depending on two parameters. We show that a reliable characterization is achievable by carefully selecting an appropriate network training strategy. These results show the viability of this approach as an effective tool for the multiparameter calibration of sensors characterized by complex transduction functions. Furthermore, the approach is proven to be versatile and promising for mass production, as the same neural network is able to calibrate different devices that have the same structure.

The characterization of quantum features in large Hilbert spaces is a crucial requirement for testing quantum protocols. In the continuous variable encoding, quantum homodyne tomography requires an amount of measurement that increases exponentially with the number of involved modes, which practically makes the protocol intractable even with few modes. Here, we introduce a new technique, based on a machine learning protocol with artificial neural networks, that allows us to directly detect negativity of the Wigner function for multimode quantum states. We test the procedure on a whole class of numerically simulated multimode quantum states for which the Wigner function is known analytically. We demonstrate that the method is fast, accurate, and more robust than conventional methods when limited amounts of data are available. Moreover, the method is applied to an experimental multimode quantum state, for which an additional test of resilience to losses is carried out.

Relevant metrological scenarios involve the simultaneous estimation of multiple parameters. The fundamental ingredient to achieve quantum-enhanced performance is based on the use of appropriately tailored quantum probes. However, reaching the ultimate resolution allowed by physical laws requires nontrivial estimation strategies from both a theoretical and a practical point of view. A crucial tool for this purpose is the application of adaptive learning techniques. Indeed, adaptive strategies provide a flexible approach to obtain optimal parameter-independent performance and optimize convergence to the fundamental bounds with a limited amount of resources. Here, we combine on the same platform quantum-enhanced multiparameter estimation attaining the corresponding quantum limit and adaptive techniques. We demonstrate the simultaneous estimation of three optical phases in a programmable integrated photonic circuit, in the limited-resource regime. The obtained results show the possibility of successfully combining different fundamental methodologies towards transition to quantum sensors applications.