arXiv subjects

Xuan Liu

Publications and source records attributed to Xuan Liu.

At least 145 records · Page 8Linked to original sources

arXiv2019

Pressure-controlled Structural Symmetry Transition in Layered InSe

Structural symmetry of crystals plays important roles in physical properties of two-dimensional (2D) materials, particularly in the nonlinear optics regime. It has been a long-term exploration on the physical properties in 2D materials with various stacking structures, which correspond to different structural symmetries. Usually, the manipulation of rotational alignment between layers in 2D heterostructures has been realized at the synthetic stage through artificial stacking like assembling Lego bricks. However, the reconfigurable control of translational symmetry of crystalline structure is still challenging. High pressure, as a powerful external control knob, provides a very promising route to circumvent this constraint. Here, we experimentally demonstrate a pressure-controlled symmetry transition in layered InSe. The continuous and reversible evolution of structural symmetries can be in-situ monitored by using the polarization-resolved second harmonic generation (SHG) spectroscopy. As pressure changes, the reconfigurable symmetry transition of the SHG pattern from three-fold rotational symmetry to mirror symmetry was experimentally observed in a layered InSe samples and was successfully explained by the proposed interlayer-translation model. This opens new routes towards potential applications of manipulating crystal symmetry of 2D materials.

physics.optics↗

arXiv2019

A uniform approach to soliton cellular automata using rigged configurations

For soliton cellular automata, we give a uniform description and proofs of the solitons, the scattering rule of two solitons, and the phase shift using rigged configurations in a number of special cases. In particular, we prove these properties for the soliton cellular automata using $B^{r,1}$ when $r$ is adjacent to $0$ in the Dynkin diagram or there is a Dynkin diagram automorphism sending $r$ to $0$.

math.CO↗

arXiv2018

Improving the Interpretability of Deep Neural Networks with Knowledge Distillation

Deep Neural Networks have achieved huge success at a wide spectrum of applications from language modeling, computer vision to speech recognition. However, nowadays, good performance alone is not sufficient to satisfy the needs of practical deployment where interpretability is demanded for cases involving ethics and mission critical applications. The complex models of Deep Neural Networks make it hard to understand and reason the predictions, which hinders its further progress. To tackle this problem, we apply the Knowledge Distillation technique to distill Deep Neural Networks into decision trees in order to attain good performance and interpretability simultaneously. We formulate the problem at hand as a multi-output regression problem and the experiments demonstrate that the student model achieves significantly better accuracy performance (about 1\% to 5\%) than vanilla decision trees at the same level of tree depth. The experiments are implemented on the TensorFlow platform to make it scalable to big datasets. To the best of our knowledge, we are the first to distill Deep Neural Networks into vanilla decision trees on multi-class datasets.

cs.LG↗

arXiv2018

Single shot time-resolved magnetic x-ray absorption at a Free Electron Laser

Ultrafast dynamics are generally investigated using stroboscopic pump-probe measurements, which characterize the sample properties for a single, specific time delay. These measurements are then repeated for a series of discrete time delays to reconstruct the overall time trace of the process. As a consequence, this approach is limited to the investigation of fully reversible phenomena. We recently introduced an off-axis zone plate based X-ray streaking technique, which overcomes this limitation by sampling the relaxation dynamics with a single femtosecond X-ray pulse streaked over a picosecond long time window. In this article we show that the X-ray absorption cross section can be employed as the contrast mechanism in this novel technique. We show that changes of the absorption cross section on the percent level can be resolved with this method. To this end we measure the ultrafast magnetization dynamics in CoDy alloy films. Investigating different chemical compositions and infrared pump fluences, we demonstrate the routine applicability of this technique. Probing in transmission the average magnetization dynamics of the entire film, our experimental findings indicate that the demagnetization time is independent of the specific infrared laser pump fluence. These results pave the way for the investigation of irreversible phenomena in a wide variety of scientific areas.

cond-mat.mtrl-sci↗

arXiv2018

Parabolic type equations associated with the Dirichlet form on the Sierpinski gasket

By using the analytic tools of Dirichlet forms, we initiate a study of some non-linear parabolic equations on Sierpinski gasket, motivated by modellings of fluid flows along a fractal (which can be considered as a simplified rough porous medium). Unlike the regular spaces case, such equation involving a convection term must take a quite different form and the convection term must be singular to the "linear part" which is determined the heat semigroup. In order to study these parabolic type equations, a new kind of Sobolev inequalities for the Dirichlet form on the gasket will be established as an effective tool for our study. These Sobolev inequalities, which are interesting by their own and in contrast to the case of Euclidean spaces, involve two $L^p$ norms with respect two mutually singular measures. By examining properties of a singular convolution of the associated heat semigroup, we derive the space-time regularity of solutions to the parabolic equations under a few technical conditions. The Burgers equations on the Sierpinski gasket are also studied, for which a maximum principle for solutions is derived using techniques from backward stochastic differential equations, and the existence and uniqueness as well as the regularity of solutions are obtained.

math.CA↗

arXiv2018

A stochastic Pontryagin maximum principle on the Sierpinski gasket

In this paper, we consider stochastic control problems on the Sierpinski gasket. An order comparison lemma is derived using heat kernel estimate for Brownian motion on the gasket. Using the order comparison lemma and techniques of BSDEs, we establish a Pontryagin stochastic maximum principle for these control problems. It turns out that the stochastic maximum principle on the Sierpinski gasket involves two necessity equations in contrast to its counterpart on Euclidean spaces. This effect is due to singularity between the Hausdorff measure and the energy dominant measure on the gasket, which is a common feature shared by many fractal spaces. The linear regulator problems on the gasket is also considered as an example.

math.PR↗

arXiv2018

Interpretable Deep Convolutional Neural Networks via Meta-learning

Model interpretability is a requirement in many applications in which crucial decisions are made by users relying on a model's outputs. The recent movement for "algorithmic fairness" also stipulates explainability, and therefore interpretability of learning models. And yet the most successful contemporary Machine Learning approaches, the Deep Neural Networks, produce models that are highly non-interpretable. We attempt to address this challenge by proposing a technique called CNN-INTE to interpret deep Convolutional Neural Networks (CNN) via meta-learning. In this work, we interpret a specific hidden layer of the deep CNN model on the MNIST image dataset. We use a clustering algorithm in a two-level structure to find the meta-level training data and Random Forest as base learning algorithms to generate the meta-level test data. The interpretation results are displayed visually via diagrams, which clearly indicates how a specific test instance is classified. Our method achieves global interpretation for all the test instances without sacrificing the accuracy obtained by the original deep CNN model. This means our model is faithful to the deep CNN model, which leads to reliable interpretations.

cs.LG↗

arXiv2018

A Graph Traversal Based Approach to Answer Non-Aggregation Questions Over DBpedia

We present a question answering system over DBpedia, filling the gap between user information needs expressed in natural language and a structured query interface expressed in SPARQL over the underlying knowledge base (KB). Given the KB, our goal is to comprehend a natural language query and provide corresponding accurate answers. Focusing on solving the non-aggregation questions, in this paper, we construct a subgraph of the knowledge base from the detected entities and propose a graph traversal method to solve both the semantic item mapping problem and the disambiguation problem in a joint way. Compared with existing work, we simplify the process of query intention understanding and pay more attention to the answer path ranking. We evaluate our method on a non-aggregation question dataset and further on a complete dataset. Experimental results show that our method achieves best performance compared with several state-of-the-art systems.

cs.CL↗

arXiv2018

Planar Cassegrain-type Schwarzschild Objective with Optical Metasurfaces

Curved reflective mirrors play an indispensable role in widely-employed Cassegrain-type optics, which consequently are inevitably bulky, and thus neither integratable nor well extendable. Through substituting geometric phase based reflective metasurfaces for curved mirrors, here a planar Cassegrain-type Schwarzschild objective is obtained, with both the focusing and imaging functionalities experimentally verified. The planar reflective objective achieved is intrinsically free of residual light and thus imposes less stringent restrictions on illuminating light and nanofabrications compared to its refractive counterpart. Our planar Cassegrain-type designs can potentially reshape and simplify the core components of the microscope and telescope systems working at a broadband spectral range, and thus opens new avenues for designing compact imaging systems.

physics.optics↗

arXiv2018

DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applications

This paper introduces DuReader, a new large-scale, open-domain Chinese ma- chine reading comprehension (MRC) dataset, designed to address real-world MRC. DuReader has three advantages over previous MRC datasets: (1) data sources: questions and documents are based on Baidu Search and Baidu Zhidao; answers are manually generated. (2) question types: it provides rich annotations for more question types, especially yes-no and opinion questions, that leaves more opportunity for the research community. (3) scale: it contains 200K questions, 420K answers and 1M documents; it is the largest Chinese MRC dataset so far. Experiments show that human performance is well above current state-of-the-art baseline systems, leaving plenty of room for the community to make improvements. To help the community make these improvements, both DuReader and baseline systems have been posted online. We also organize a shared competition to encourage the exploration of more models. Since the release of the task, there are significant improvements over the baselines.

cs.CL↗

arXiv2017

Backward problems for stochastic differential equations on the Sierpinski gasket

In this paper, we study the non-linear backward problems (with deterministic or stochastic durations) of stochastic differential equations on the Sierpinski gasket. We prove the existence and uniqueness of solutions of backward stochastic differential equations driven by Brownian martingale (defined in Section [sec:-1]) on the Sierpinski gasket constructed by S. Goldstein and S. Kusuoka. The exponential integrability of quadratic processes for martingale additive functionals is obtained, and as an application, a Feynman-Kac representation formula for weak solutions of semi-linear parabolic PDEs on the gasket is also established.

math.PR↗

arXiv2017

Sobolev inequalities on product Sierpinski spaces

On fractals, different measures (mutually singular in general) are involved to measure volumes of sets and energies of functions. Singularity of measures brings difficulties in (especially non-linear) analysis on fractals. In this paper, we prove a type of Sobolev inequalities, which involve different and possibly mutually singular measures, on product Sierpinski spaces. Sufficient and necessary conditions for the validity of these Sobolev inequalities are given. Furthermore, we compute the sharp exponents which appears in the sufficient and necessary conditions for the product Kusuoka measure, i.e. the reference energy measure on Sierpinski spaces.

math.CA↗

arXiv2016

On the maximum of a type of random processes

We consider a type of random processes which satisfies the conditional increment condition and obtain an estimate for the tail probability and a Doob-type inequality of the maximum of the process. The main result is that, for processes satisfying the conditional increment condition, the tail probability decay of its maximum behaves in the same way as its marginals.

math.PR↗

arXiv2016

A Decomposition Based Approach for Solving a General Bilevel Linear Programming

Bilevel optimization has been widely used in decision-making process. However, there still lacks an efficient algorithm to determine an optimal solution of a bilevel optimization problem, especially for a large-size problem. To bridge the gap, this paper proposes an efficient decomposition algorithm for a general bilevel linear programming(GBLP). The simulation results on large-size testing system demonstrate its correctness and efficiency.

math.OC↗

arXiv2016

A Quasi-sure Non-degeneracy Property for the Brownian Rough Path

In the present paper, we are going to show that outside a slim set in the sense of Malliavin (or quasi-surely), the signature path (which consists of iterated path integrals in every degree) of Brownian motion is non-self-intersecting. This property relates closely to a non-degeneracy property for the Brownian rough path arising naturally from the uniqueness of signature problem in rough path theory. As an important consequence we conclude that quasi-surely, the Brownian rough path does not have any tree-like pieces and every sample path of Brownian motion is uniquely determined by its signature up to reparametrization.

math.PR↗

arXiv2015

Impacts of Bad Data on the PMU based Line Outage Detection

Power systems become more prone to cyber-attacks due to the high integration of information technologies. In this paper, we demonstrate that the outages of some lines can be masked by injecting false data into a set of measurements. The success of the topology attack can be guaranteed by making that: 1)the injected false data obeys KCL and KVL to avoid being detected by the bad data detection program in the state estimation; 2)the residual is increased such that the line outage cannot be detected by PMU data. A quadratic programming problem is set up to determine the optimal attack vector that can maximize the residual of the outaged line. The IEEE 39-bus system is used to demonstrate the masking scheme.

cs.CR↗

arXiv2013

LED-Induced Fluorescence System for Tea Classification and Quality Assessment

A fluorescence system is developed by using several light emitting diodes (LEDs) with different wavelengths as excitation light sources. The fluorescence detection head consists of multi LED light sources and a multimode fiber for fluorescence collection, where the LEDs and the corresponding filters can be easily chosen to get appropriate excitation wavelengths for different applications. By analyzing fluorescence spectra with the principal component analysis method, the system is utilized in the classification of four types of green tea beverages and two types of black tea beverages. Qualities of the Xihu Longjing tea leaves of different grades, as well as the corresponding liquid tea samples, are studied to further investigate the ability and application of the system in the evaluation of classification/quality of tea and other foods.

physics.optics↗

arXiv2013

A generalized theory of preferential linking

There are diverse mechanisms driving the evolution of social networks. A key open question dealing with understanding their evolution is: How various preferential linking mechanisms produce networks with different features? In this paper we first empirically study preferential linking phenomena in an evolving online social network, find and validate the linear preference. We propose an analyzable model which captures the real growth process of the network and reveals the underlying mechanism dominating its evolution. Furthermore based on preferential linking we propose a generalized model reproducing the evolution of online social networks, present unified analytical results describing network characteristics for 27 preference scenarios, and explore the relation between preferential linking mechanism and network features. We find that within the framework of preferential linking analytical degree distributions can only be the combinations of finite kinds of functions which are related to rational, logarithmic and inverse tangent functions, and extremely complex network structure will emerge even for very simple sublinear preferential linking. This work not only provides a verifiable origin for the emergence of various network characteristics in social networks, but bridges the micro individuals' behaviors and the global organization of social networks.

physics.soc-ph↗