No abstract

It is a fundamental challenge in quantum optics to deterministically generate indistinguishable single photons through non-deterministic nonlinear optical processes, due to the intrinsic coupling of single- and multi-photon-generation probabilities in these processes. Actively multiplexing photons generated in many temporal modes can decouple these probabilities, but key issues are to minimize resource requirements to allow scalability, and to ensure indistinguishability of the generated photons. Here we demonstrate the multiplexing of photons from four temporal modes solely using fibre-integrated optics and off-the-shelf electronic components. We show a 100% enhancement to the single-photon output probability without introducing additional multi-photon noise. Photon indistinguishability is confirmed by a fourfold Hong–Ou–Mandel quantum interference with a 91±16% visibility after subtracting multi-photon noise due to high pump power. Our demonstration paves the way for scalable multiplexing of many non-deterministic photon sources to a single near-deterministic source, which will be of benefit to future quantum photonic technologies.

Rapid generation of high quality Gaussian random numbers is a key capability for simulations across a wide range of disciplines. Advances in computing have brought the power to conduct simulations with very large numbers of random numbers and with it, the challenge of meeting increasingly stringent requirements on the quality of Gaussian random number generators (GRNG). This article describes the algorithms underlying various GRNGs, compares their computational requirements, and examines the quality of the random numbers with emphasis on the behaviour in the tail region of the Gaussian probability density function.

Inference for state-of-the-art deep neural networks is computationally expensive, making them difficult to deploy on constrained hardware environments. An efficient way to reduce this complexity is to quantize the weight parameters and/or activations during training by approximating their distributions with a limited entry codebook. For very low-precisions, such as binary or ternary networks with 1-8-bit activations, the information loss from quantization leads to significant accuracy degradation due to large gradient mismatches between the forward and backward functions. In this paper, we introduce a quantization method to reduce this loss by learning a symmetric codebook for particular weight subgroups. These subgroups are determined based on their locality in the weight matrix, such that the hardware simplicity of the low-precision representations is preserved. Empirically, we show that symmetric quantization can substantially improve accuracy for networks with extremely low-precision weights and activations. We also demonstrate that this representation imposes minimal or no hardware implications to more coarse-grained approaches. Source code is available at https://www.github.com/julianfaraone/SYQ.

No abstract