Project description
New approach for testing quantum supremacy
To test the ability of quantum computing devices to solve problems that classical computers practically cannot, researchers have looked into the idea of Gaussian boson sampling. This works by creating an environment where photons are introduced into a device and allowed to interact with each other for a given period of time. Prior research suggested that this is an intractable problem for todays’ computers as simulating photon location distribution over multiple samples would take a very long time. The EU-funded FrEQuMP project will conduct boson sampling by considering the frequency information encoded in photonic quantum states. They can then generate the desired coherent quantum state and demonstrate boson sampling with many photons, avoiding problems arising from photons leaking from the system.
Objective
Optical quantum computing and quantum simulation rely on multi-photon interference effects, between many photons in a larger number of optical paths or modes. In particular in Gaussian Boson Sampling protocols, single-mode squeezed states are input to an optical circuit implementing a transformation on the modes, which creates a complex multi-mode squeezed state. Sampling from such a state with single photon detectors is thought to be an intractable problem to simulate with classical computers, and has useful applications, for instance in calculating molecular vibronic spectra and in identifying densely connected sub-graphs of a network. This motivates building quantum-optical devices to implement Gaussian Boson Sampling. However, it is resource intensive to create a usefully large state using many separate squeezed sources and a circuit, and very technically challenging to avoid photon loss and to maintain interferometric stability. Here, I propose to carry out Gaussian Boson Sampling and related experiments by directly generating multi-mode squeezed states encoded in frequency, from a single source with reconfigurable frequency correlations. Using frequency channels to represent the modes is very compact because they can all propagate along the same spatial path, and this also ensures interferometric stability. Directly generating the desired state will avoid having the photons propagate through a lossy circuit, allowing scaling to higher photon numbers, and frequency encoding will make large numbers of modes readily available, surpassing the state-of-the-art in spatially-encoded circuits.
Fields of science
Keywords
Programme(s)
Funding Scheme
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinator
SW7 2AZ LONDON
United Kingdom