arXiv subjects

Si Chen

Publications and source records attributed to Si Chen.

At least 109 records · Page 6Linked to original sources

arXiv2021

Sequential Recommendation in Online Games with Multiple Sequences, Tasks and User Levels

Online gaming is growing faster than ever before, with increasing challenges of providing better user experience. Recommender systems (RS) for online games face unique challenges since they must fulfill players' distinct desires, at different user levels, based on their action sequences of various action types. Although many sequential RS already exist, they are mainly single-sequence, single-task, and single-user-level. In this paper, we introduce a new sequential recommendation model for multiple sequences, multiple tasks, and multiple user levels (abbreviated as M$^3$Rec) in Tencent Games platform, which can fully utilize complex data in online games. We leverage Graph Neural Network and multi-task learning to design M$^3$Rec in order to model the complex information in the heterogeneous sequential recommendation scenario of Tencent Games. We verify the effectiveness of M$^3$Rec on three online games of Tencent Games platform, in both offline and online evaluations. The results show that M$^3$Rec successfully addresses the challenges of recommendation in online games, and it generates superior recommendations compared with state-of-the-art sequential recommendation approaches.

cs.AI↗

arXiv2021

Self-Amplification of Coherent Energy Modulation in Seeded Free-Electron Lasers

The spectroscopic techniques for time-resolved fine analysis of matter require coherent X-ray radiation with femtosecond duration and high average brightness. Seeded free-electron lasers (FELs), which use the frequency up-conversion of an external seed laser to improve temporal coherence, are ideal for providing fully coherent soft X-ray pulses. However, it is difficult to operate seeded FELs at a high repetition rate due to the limitations of present state-of-the-art laser systems. Here, we report the novel self-modulation method for enhancing laser-induced energy modulation, thereby significantly reducing the requirement of an external laser system. Driven by this scheme, we experimentally realize high harmonic generation in a seeded FEL using an unprecedentedly small energy modulation. An electron beam with a laser-induced energy modulation as small as 1.8 times the slice energy spread is used for lasing at the 7th harmonic of a 266-nm seed laser in a single-stage high-gain harmonic generation (HGHG) setup and the 30th harmonic of the seed laser in a two-stage HGHG setup. The results mark a major step towards a high-repetition-rate, fully coherent X-ray FEL.

physics.acc-ph↗

arXiv2020

Resolution enhancement and realistic speckle recovery with generative adversarial modeling of micro-optical coherence tomography

A resolution enhancement technique for optical coherence tomography (OCT), based on Generative Adversarial Networks (GANs), was developed and investigated. GANs have been previously used for resolution enhancement of photography and optical microscopy images. We have adapted and improved this technique for OCT image generation. Conditional GANs (cGANs) were trained on a novel set of ultrahigh resolution spectral domain OCT volumes, termed micro-OCT, as the high-resolution ground truth (~1$μ$m isotropic resolution). The ground truth was paired with a low-resolution image obtained by synthetically degrading resolution 4x in one of (1-D) or both axial and lateral axes (2-D). Cross-sectional image (B-scan) volumes obtained from in vivo imaging of human labial (lip) tissue and mouse skin were used in separate feasibility experiments. Accuracy of resolution enhancement compared to ground truth was quantified with human perceptual accuracy tests performed by an OCT expert. The GAN loss in the optimization objective, noise injection in both the generator and discriminator models, and multi-scale discrimination were found to be important for achieving realistic speckle appearance in the generated OCT images. The utility of high resolution speckle recovery was illustrated by an example of micro-OCT imaging of blood vessels in lip tissue. Qualitative examples applying the models to image data from outside of the training data distribution, namely human retina and mouse bladder, were also demonstrated, suggesting potential for cross-domain transferability. This preliminary study suggests that deep learning generative models trained on OCT images from high-performance prototype systems may have potential in enhancing lower resolution data from mainstream/commercial systems, thereby bringing cutting-edge technology to the masses at low cost.

eess.IV↗

arXiv2020

Observation of intensity squeezing in resonance fluorescence from a solid-state device

Intensity squeezing, i.e., photon number fluctuations below the shot noise limit, is a fundamental aspect of quantum optics and has wide applications in quantum metrology. It was predicted in 1979 that the intensity squeezing could be observed in resonance fluorescence from a two-level quantum system. Yet, its experimental observation in solid states was hindered by inefficiencies in generating, collecting and detecting resonance fluorescence. Here, we report the intensity squeezing in a single-mode fibre-coupled resonance fluorescence single-photon source based on a quantum dot-micropillar system. We detect pulsed single-photon streams with 22.6% system efficiency, which show subshot-noise intensity fluctuation with an intensity squeezing of 0.59 dB. We estimate a corrected squeezing of 3.29 dB at the first lens. The observed intensity squeezing provides the last piece of the fundamental picture of resonance fluorescence; which can be used as a new standard for optical radiation and in scalable quantum metrology with indistinguishable single photons.

quant-ph↗

arXiv2020

Attacking Recommender Systems with Augmented User Profiles

Recommendation Systems (RS) have become an essential part of many online services. Due to its pivotal role in guiding customers towards purchasing, there is a natural motivation for unscrupulous parties to spoof RS for profits. In this paper, we study the shilling attack: a subsistent and profitable attack where an adversarial party injects a number of user profiles to promote or demote a target item. Conventional shilling attack models are based on simple heuristics that can be easily detected, or directly adopt adversarial attack methods without a special design for RS. Moreover, the study on the attack impact on deep learning based RS is missing in the literature, making the effects of shilling attack against real RS doubtful. We present a novel Augmented Shilling Attack framework (AUSH) and implement it with the idea of Generative Adversarial Network. AUSH is capable of tailoring attacks against RS according to budget and complex attack goals, such as targeting a specific user group. We experimentally show that the attack impact of AUSH is noticeable on a wide range of RS including both classic and modern deep learning based RS, while it is virtually undetectable by the state-of-the-art attack detection model.

cs.IR↗

arXiv2020

Object-Adaptive LSTM Network for Real-time Visual Tracking with Adversarial Data Augmentation

In recent years, deep learning based visual tracking methods have obtained great success owing to the powerful feature representation ability of Convolutional Neural Networks (CNNs). Among these methods, classification-based tracking methods exhibit excellent performance while their speeds are heavily limited by the expensive computation for massive proposal feature extraction. In contrast, matching-based tracking methods (such as Siamese networks) possess remarkable speed superiority. However, the absence of online updating renders these methods unadaptable to significant object appearance variations. In this paper, we propose a novel real-time visual tracking method, which adopts an object-adaptive LSTM network to effectively capture the video sequential dependencies and adaptively learn the object appearance variations. For high computational efficiency, we also present a fast proposal selection strategy, which utilizes the matching-based tracking method to pre-estimate dense proposals and selects high-quality ones to feed to the LSTM network for classification. This strategy efficiently filters out some irrelevant proposals and avoids the redundant computation for feature extraction, which enables our method to operate faster than conventional classification-based tracking methods. In addition, to handle the problems of sample inadequacy and class imbalance during online tracking, we adopt a data augmentation technique based on the Generative Adversarial Network (GAN) to facilitate the training of the LSTM network. Extensive experiments on four visual tracking benchmarks demonstrate the state-of-the-art performance of our method in terms of both tracking accuracy and speed, which exhibits great potentials of recurrent structures for visual tracking.

cs.CV↗

arXiv2020

Adaptive Deep Metric Embeddings for Person Re-Identification under Occlusions

Person re-identification (ReID) under occlusions is a challenging problem in video surveillance. Most of existing person ReID methods take advantage of local features to deal with occlusions. However, these methods usually independently extract features from the local regions of an image without considering the relationship among different local regions. In this paper, we propose a novel person ReID method, which learns the spatial dependencies between the local regions and extracts the discriminative feature representation of the pedestrian image based on Long Short-Term Memory (LSTM), dealing with the problem of occlusions. In particular, we propose a novel loss (termed the adaptive nearest neighbor loss) based on the classification uncertainty to effectively reduce intra-class variations while enlarging inter-class differences within the adaptive neighborhood of the sample. The proposed loss enables the deep neural network to adaptively learn discriminative metric embeddings, which significantly improve the generalization capability of recognizing unseen person identities. Extensive comparative evaluations on challenging person ReID datasets demonstrate the significantly improved performance of the proposed method compared with several state-of-the-art methods.

cs.CV↗

arXiv2020

Joint Deep Learning of Facial Expression Synthesis and Recognition

Recently, deep learning based facial expression recognition (FER) methods have attracted considerable attention and they usually require large-scale labelled training data. Nonetheless, the publicly available facial expression databases typically contain a small amount of labelled data. In this paper, to overcome the above issue, we propose a novel joint deep learning of facial expression synthesis and recognition method for effective FER. More specifically, the proposed method involves a two-stage learning procedure. Firstly, a facial expression synthesis generative adversarial network (FESGAN) is pre-trained to generate facial images with different facial expressions. To increase the diversity of the training images, FESGAN is elaborately designed to generate images with new identities from a prior distribution. Secondly, an expression recognition network is jointly learned with the pre-trained FESGAN in a unified framework. In particular, the classification loss computed from the recognition network is used to simultaneously optimize the performance of both the recognition network and the generator of FESGAN. Moreover, in order to alleviate the problem of data bias between the real images and the synthetic images, we propose an intra-class loss with a novel real data-guided back-propagation (RDBP) algorithm to reduce the intra-class variations of images from the same class, which can significantly improve the final performance. Extensive experimental results on public facial expression databases demonstrate the superiority of the proposed method compared with several state-of-the-art FER methods.

cs.CV↗

arXiv2019

Boson sampling with 20 input photons in 60-mode interferometers at $10^{14}$ state spaces

Quantum computing experiments are moving into a new realm of increasing size and complexity, with the short-term goal of demonstrating an advantage over classical computers. Boson sampling is a promising platform for such a goal, however, the number of involved single photons was up to five so far, limiting these small-scale implementations to a proof-of-principle stage. Here, we develop solid-state sources of highly efficient, pure and indistinguishable single photons, and 3D integration of ultra-low-loss optical circuits. We perform an experiment with 20 single photons fed into a 60-mode interferometer, and, in its output, sample over Hilbert spaces with a size of $10^{14}$ $-$over ten orders of magnitude larger than all previous experiments. The results are validated against distinguishable samplers and uniform samplers with a confidence level of 99.9%.

quant-ph↗

arXiv2019

FS-indicators of pq-dimensional pointed Hopf algebras

We compute higher Frobenius-Schur indicators of pq-dimensional pointed Hopf algebras in characteristic p through their associated graded Hopf algebras. These indicators are gauge invariants for the monoidal categories of representations of these algebras.

math.QA↗

arXiv2019

Towards optimal single-photon sources from polarized microcavities

An optimal single-photon source should deterministically deliver one and only one photon at a time, with no trade-off between the source's efficiency and the photon indistinguishability. However, all reported solid-state sources of indistinguishable single photons had to rely on polarization filtering which reduced the efficiency by 50%, which fundamentally limited the scaling of photonic quantum technologies. Here, we overcome this final long-standing challenge by coherently driving quantum dots deterministically coupled to polarization-selective Purcell microcavities--two examples are narrowband, elliptical micropillars and broadband, elliptical Bragg gratings. A polarization-orthogonal excitation-collection scheme is designed to minimize the polarization-filtering loss under resonant excitation. We demonstrate a polarized single-photon efficiency of 0.60+/-0.02 (0.56+/-0.02), a single-photon purity of 0.975+/-0.005 (0.991+/-0.003), and an indistinguishability of 0.975+/-0.006 (0.951+/-0.005) for the micropillar (Bragg grating) device. Our work provides promising solutions for truly optimal single-photon sources combining near-unity indistinguishability and near-unity system efficiency simultaneously.

quant-ph↗

arXiv2019

Contrast of nuclei in stratified squamous epithelium in optical coherence tomography images at 800 nm

Imaging nuclei of keratinocytes in the stratified squamous epithelium has been a subject of intense research since nucleus associated cellular atypia is the key criteria for the screening and diagnosis of epithelial cancers and their precursors. However, keratinocyte nuclei have been reported to be either low scattering or high scattering, so that these inconsistent reports might have led to misinterpretations of optical images, and more importantly, hindered the establishment of optical diagnostic criteria. We disclose that they are generally low scattering in the core using Micro-optical coherence tomography (micro-OCT) of 1.28 um axial resolution in vivo; those previously reported high scattering or bright signals from nuclei are likely from the nucleocytoplasmic boundary, and the low-scattering nuclear cores were missed possibly due to insufficient axial resolutions (about 4 um). It is further demonstrated that the high scattering signals may be associated with flattening of nuclei and cytoplasmic glycogen accumulation, which are valuable cytologic hallmarks of cell maturation.

q-bio.TO↗

arXiv2019

Coherently driving a single quantum two-level system with dichromatic laser pulses

Efficient excitation of a single two-level system usually requires that the driving field is at the same frequency as the atomic transition. However, the scattered laser light in solid-state implementations can dominate over the single photons, imposing an outstanding challenge to perfect single-photon sources. Here, we propose a background-free method using a phase-locked dichromatic electromagnetic field with no spectral overlap with the optical transition for a coherent control of a two-level system, and we demonstrate this method experimentally with a single quantum dot embedded in a micropillar. Single photons generated by pi excitation show a purity of 0.988(1) and indistinguishability of 0.962(6). Further, the phase-coherent nature of the two-color excitation is captured by the resonance-fluorescence intensity dependence on the relative phase between the two pulses. Our two-color excitation method adds a useful toolbox to the study of atom-photon interaction, and the generation of spectrally isolated indistinguishable single photons.

quant-ph↗

arXiv2019

The Global Convergence Analysis of the Bat Algorithm Using a Markovian Framework and Dynamical System Theory

The bat algorithm (BA) has been shown to be effective to solve a wider range of optimization problems. However, there is not much theoretical analysis concerning its convergence and stability. In order to prove the convergence of the bat algorithm, we have built a Markov model for the algorithm and proved that the state sequence of the bat population forms a finite homogeneous Markov chain, satisfying the global convergence criteria. Then, we prove that the bat algorithm can have global convergence. In addition, in order to enhance the convergence performance of the algorithm, we have designed an updated model using the dynamical system theory in terms of a dynamic matrix, and the parameter ranges for the algorithm stability are then obtained. We then use some benchmark functions to demonstrate that BA can indeed achieve global optimality efficiently for these functions.

math.OC↗

arXiv2019

Polarization constraints in reciprocal unitary backscattering

We observed that the polarization state of light after round-trip propagation through a birefringent medium frequently aligns with the employed input polarization state "mirrored" by the horizontal plane of the Poincare sphere. In this letter we explore the predisposition for this mirror state and demonstrate how it constrains the evolution of polarization states as a function of the round-trip depth into weakly scattering birefringent samples, as measured with polarization-sensitive optical coherence tomography (PS-OCT). The constraint enables measurements of depth-resolved sample birefringence with PS-OCT using only a single input polarization state, which offers a critical simplification compared to the use of multiple input states.

physics.optics↗

arXiv2018

Temperature and electric field dependence of spin relaxation in graphene on SrTiO$_3$

The theoretically predicted intrinsic spin relaxation time of up to 1 $μs$ in graphene along with extremely high mobilities makes it a promising material in spintronics. In spite of extensive experimental studies of spin relaxation and understanding of its precise mechanism, it is still unclear as to why the spin lifetime in graphene is three orders of magnitude below the theoretical predictions. Central to this discrepancy is the role of the local environment including that of the underlying substrate. In this work, we use the electronically rich platform SrTiO$_3$ and study its suitability in supporting spin transport in graphene. We find spin relaxation time and length as large as 1.2 $\pm$ 0.1 ns and 5.6 $\pm$ 0.5 $μm$ respectively at 290 K in graphene on SrTiO$_3$ using a non-local measurement scheme. We analyze the temperature variation of the spin transport parameters in graphene and attribute the temperature dependence of the spin transport parameters in graphene to spin orbit coupling or structural phase transition in SrTiO$_3$. Furthermore, from the gate dependence of the spin transport parameters, the relation between spin relaxation time and momentum relaxation time is extracted; the Elliot-Yaffet and D'Yakonov-Perel' spin relaxation rates are found to be of similar order.

cond-mat.mes-hall↗

arXiv2018

Polarized indistinguishable single photons from a quantum dot in an elliptical micropillar

The key challenge to scalable optical quantum computing, boson sampling, and quantum metrology is sources of single photons with near-unity system efficiency and simultaneously near-perfect indistinguishability in all degrees of freedom (including spectral, temporal, spatial, and polarization). However, previous high-indistinguishability solid-state single-photon sources had to rely on polarization filtering that reduced the system efficiency by at least 50%. Here, we overcome this challenge by developing a new single-photon source based on a coherently driven quantum dot embedded in an elliptical micropillar. The asymmetric cavity lifts the polarization degeneracy into two orthogonal linearly polarized modes with a suitable energy separation. We design an excitation-collection scheme that allows the creation and collection of single photons with an indistinguishability of 0.976(1) and a degree of polarization of 91%. Our method provides a solution of combining near-unity system efficiency and indistinguishability compatible with background-free resonant excitation, and opens the way to truly optimal single-photon sources for scalable photonic quantum technologies.

physics.optics↗

arXiv2018

Multi-task Learning of Cascaded CNN for Facial Attribute Classification

Recently, facial attribute classification (FAC) has attracted significant attention in the computer vision community. Great progress has been made along with the availability of challenging FAC datasets. However, conventional FAC methods usually firstly pre-process the input images (i.e., perform face detection and alignment) and then predict facial attributes. These methods ignore the inherent dependencies among these tasks (i.e., face detection, facial landmark localization and FAC). Moreover, some methods using convolutional neural network are trained based on the fixed loss weights without considering the differences between facial attributes. In order to address the above problems, we propose a novel multi-task learning of cas- caded convolutional neural network method, termed MCFA, for predicting multiple facial attributes simultaneously. Specifically, the proposed method takes advantage of three cascaded sub-networks (i.e., S_Net, M_Net and L_Net corresponding to the neural networks under different scales) to jointly train multiple tasks in a coarse-to-fine manner, which can achieve end-to-end optimization. Furthermore, the proposed method automatically assigns the loss weight to each facial attribute based on a novel dynamic weighting scheme, thus making the proposed method concentrate on predicting the more difficult facial attributes. Experimental results show that the proposed method outperforms several state-of-the-art FAC methods on the challenging CelebA and LFWA datasets.

cs.CV↗