Xianxin Guo

Xianxin Guo

Auto-generated profile, based on publicly available information

Xianxin Guo

Department of Physics, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 August 2017 in Physical Review Letters
Verified
Journal Article
Closed access
PDF file
YZYueyang Zou
Yueyang Zou
YJYue Jiang
Yue Jiang
YMYefeng Mei
Yefeng Mei
XGXianxin Guo
Xianxin Guo
Shengwang DuShengwang Du
Shengwang Du
We report an experiment demonstrating the generation of directional thermal radiation with a spectral brightness that is about 9 times greater than that of the ambient pumping reservoir. The experiment is based on the recent proposal for a nontraditional quantum heat engine and uses cold Rb atoms, electromagnetically induced transparency, and photon correlation spectroscopy [ Phys. Rev. A 94, 053859 (2016)].

14 July 2022 in Optica
Verified
Journal Article
Open access
JSJames Spall
James Spall
XGXianxin Guo
Xianxin Guo
ALAlexander Lvovsky
Alexander Lvovsky
Optical neural networks are emerging as a promising type of machine learning hardware capable of energy-efficient, parallel computation. Today’s optical neural networks are mainly developed to perform optical inference after in silico training on digital simulators. However, various physical imperfections that cannot be accurately modeled may lead to the notorious “reality gap” between the digital simulator and the physical system. To address this challenge, we demonstrate hybrid training of optical neural networks where the weight matrix is trained with neuron activation functions computed optically via forward propagation through the network. We examine the efficacy of hybrid training with three different networks: an optical linear classifier, a hybrid opto-electronic network, and a complex-valued optical network. We perform a study comparative to in silico training, and our results show that hybrid training is robust against different kinds of static noise. Our platform-agnostic hybrid training scheme can be applied to a wide variety of optical neural networks, and this work paves the way towards advanced all-optical training in machine intelligence.

6 August 2014 in Optica
Verified
Journal Article
Open access
PDF file
LZLuwei Zhao
Luwei Zhao
XGXianxin Guo
Xianxin Guo
CLChang Liu
Chang Liu
YSYuan Sun
Yuan Sun
MLM. M. T. Loy
M. M. T. Loy
Shengwang DuShengwang Du
Shengwang Du
The generation of nonclassical photon pairs with a long coherence time is key for applications that range from fundamental to quantum communication and metrology. Spontaneous four-wave mixing with electromagnetically induced transparency has been demonstrated as one of the most efficient methods; however, narrowing the bandwidth and producing photon pairs with a temporal length beyond 1 μs remains a technical challenge due to noise considerations and the need for cold atoms with a high optical depth (OD). In this work, we demonstrate the generation of narrowband photon pairs with a controllable coherence time up to 1.72 μs in a laser-cooled atomic ensemble with an OD as high as 130. At such a high OD, we find that the pump laser field spatial profile has a significant effect on the time–frequency entangled two-photon waveform. We also confirm the quantum particle nature of heralded narrowband single photons generated from this source.

1 March 2021 in Photonics Research
Verified
Journal Article
Closed access
XGXianxin Guo
Xianxin Guo
TBThomas D. Barrett
Thomas D. Barrett
ZWZhiming M. Wang
Zhiming M. Wang
ALAlexander Lvovsky
Alexander Lvovsky
We propose a practical scheme for end-to-end optical backpropagation in neural networks. Using saturable absorption for the nonlinear units, we find that the backward-propagating gradients required to train the network can be approximated in a surprisingly simple pump-probe scheme that requires only simple passive optical elements. Simulations show that, with readily obtainable optical depths, our approach can achieve equivalent performance to state-of-the-art computational networks on image classification benchmarks, even in deep networks with multiple sequential gradient approximation. With backpropagation through nonlinear units being an outstanding challenge to the field, this work provides a feasible path toward truly all-optical neural networks.

17 April 2017 in Applied Physics Letters
Verified
Journal Article
Open access
PDF file
LZLingbang Zhu
Lingbang Zhu
XGXianxin Guo
Xianxin Guo
CSChi Shu
Chi Shu
HJHeejeong Jeong
Heejeong Jeong
Shengwang DuShengwang Du
Shengwang Du
We demonstrate the generation of high-quality narrowband biphotons from a Doppler-broadened hot rubidium atomic vapor cell. Choosing a double-Λ atomic energy level scheme for optimizing both spontaneous four-wave mixing nonlinear parametric interaction and electromagnetically induced transparency (EIT), we achieve a biphoton spectral brightness as high as 14 000 s−1 MHz−1. Meanwhile, we apply a spatially tailored optical pumping beam for reduction of the Raman noise and obtain a violation of the Cauchy-Schwarz inequality by a factor of 1023.

22 March 2017 in Optica
Verified
Journal Article
Open access
PDF file
XGXianxin Guo
Xianxin Guo
YMYefeng Mei
Yefeng Mei
Shengwang DuShengwang Du
Shengwang Du
Entanglement, describing the inseparability of a quantum multiparty system, is one of the most intriguing features of quantum mechanics. Violation of Bell inequality, for ruling out the possibility of local hidden-variable theories, is commonly used as a strong witness of quantum entanglement. In previous Bell test experiments with photonic entanglement based on two-photon coincidence measurement, the photon temporal wave packets were absorbed completely by the detectors. That is, the photon coherence time is much shorter than the detection time window. Here we demonstrate the generation of frequency-bin entangled narrowband biphotons, and for the first time, to the best of our knowledge, test the Clauser–Horne–Shimony–Holt (CHSH) Bell inequality |S|≤2 for their nonlocal temporal correlations with time-resolved detection. We obtain a maximum |S| value of 2.52±0.48, which violates the CHSH inequality. Our result will have applications in quantum information processing involving time-frequency entanglement.