Downloads & Free Reading Options - Results

We suggest and demonstrate a scheme for coherent nonlocal compensation of pure phase objects based on two-photon polarization and momentum entangled states. The insertion of a single phase object on one of the beams reduces the purity of the state and the amount of shared entanglement, whereas the original entanglement can be retrieved by adding a suitable phase object on the other beam. In our setup polarization and momentum entangled states are generated by spontaneous parametric downconversion and then purified using a programmable spatial light modulator, which may be also used to impose arbitrary space dependent phase functions to the beams. As a possible application, we suggest and demonstrate a quantum key distribution protocol based on nonlocal phase compensation.

We suggest and demonstrate a scheme for coherent nonlocal compensation of pure phase objects based on two-photon polarization and momentum entangled states. The insertion of a single phase object on one of the beams reduces the purity of the state and the amount of shared entanglement, whereas the original entanglement can be retrieved by adding a suitable phase object on the other beam. In our setup polarization and momentum entangled states are generated by spontaneous parametric downconversion and then purified using a programmable spatial light modulator, which may be also used to impose arbitrary space dependent phase functions to the beams. As a possible application, we suggest and demonstrate a quantum key distribution protocol based on nonlocal phase compensation.

In 1935, Einstein, Podolsky and Rosen (EPR) considered two particles in an entangled state of motion to illustrate why they questioned the completeness of quantum theory. In the past decades, microscopic systems with entanglement in various degrees of freedom have successfully been generated, representing compelling evidence to support the completeness of quantum theory. Today, the generation of an EPR-entangled state of motion of two massive objects of up to the kilogram-scale seems feasible with state-of-the-art technology. Recently, the generation and verification of EPR-entangled mirror motion in interferometric gravitational wave detectors was proposed, with the aim of testing quantum theory in the regime of macroscopic objects, and to make available nonclassical probe systems for future tests of modified quantum theories that include (non-relativistic) gravity. The work presented here builds on these earlier results and proposes a specific Michelson interferometer that includes two high-quality laser mirrors of about 0.1 kg mass each. The mirrors are individually suspended as pendula and located close to each other, and cooled to about 4 K. The physical concepts for the generation of the EPR-entangled centre of mass motion of these two mirrors are described. Apart from a test of quantum mechanics in the macroscopic world, the setup is envisioned to test predictions of yet-to-be-elaborated modified quantum theories that include gravitational effects.

Our world can be succinctly and compactly described as structured scenes of objects and relations. A typical room, for example, contains salient objects such as tables, chairs and books, and these objects typically relate to each other by their underlying causes and semantics. This gives rise to correlated features, such as position, function and shape. Humans exploit knowledge of objects and their relations for learning a wide spectrum of tasks, and more generally when learning the structure underlying observed data. In this work, we introduce relation networks (RNs) - a general purpose neural network architecture for object-relation reasoning. We show that RNs are capable of learning object relations from scene description data. Furthermore, we show that RNs can act as a bottleneck that induces the factorization of objects from entangled scene description inputs, and from distributed deep representations of scene images provided by a variational autoencoder. The model can also be used in conjunction with differentiable memory mechanisms for implicit relation discovery in one-shot learning tasks. Our results suggest that relation networks are a potentially powerful architecture for solving a variety of problems that require object relation reasoning.