arXiv subjects

Dong Wang

Publications and source records attributed to Dong Wang.

At least 793 records · Page 44Linked to original sources

arXiv2015

Stochastic Top-k ListNet

ListNet is a well-known listwise learning to rank model and has gained much attention in recent years. A particular problem of ListNet, however, is the high computation complexity in model training, mainly due to the large number of object permutations involved in computing the gradients. This paper proposes a stochastic ListNet approach which computes the gradient within a bounded permutation subset. It significantly reduces the computation complexity of model training and allows extension to Top-k models, which is impossible with the conventional implementation based on full-set permutations. Meanwhile, the new approach utilizes partial ranking information of human labels, which helps improve model quality. Our experiments demonstrated that the stochastic ListNet method indeed leads to better ranking performance and speeds up the model training remarkably.

cs.IR↗

arXiv2015

Correlation kernels for sums and products of random matrices

Let $X$ be a random matrix whose squared singular value density is a polynomial ensemble. We derive double contour integral formulas for the correlation kernels of the squared singular values of $GX$ and $TX$, where $G$ is a complex Ginibre matrix and $T$ is a truncated unitary matrix. We also consider the product of $X$ and several complex Ginibre/truncated unitary matrices. As an application, we derive the precise condition for the squared singular values of the product of several truncated unitary matrices to follow a polynomial ensemble. We also consider the sum $H + M$ where $H$ is a GUE matrix and $M$ is a random matrix whose eigenvalue density is a polynomial ensemble. We show that the eigenvalues of $H + M$ follow a polynomial ensemble whose correlation kernel can be expressed as a double contour integral. As an application, we point out a connection to the two-matrix model.

math.PR↗

arXiv2015

Fluctuations of TASEP and LPP with general initial data

We prove Airy process variational formulas for the one-point probability distribution of (discrete time parallel update) TASEP with general initial data, as well as last passage percolation from a general lattice path to a point. We also consider variants of last passage percolation with inhomogeneous parameter geometric weights and provide variational formulas of a similar nature. This proves one aspect of the conjectural description of the renormalization fixed point of the Kardar-Parisi-Zhang universality class.

math.PR↗

arXiv2015

Learning from LDA using Deep Neural Networks

Latent Dirichlet Allocation (LDA) is a three-level hierarchical Bayesian model for topic inference. In spite of its great success, inferring the latent topic distribution with LDA is time-consuming. Motivated by the transfer learning approach proposed by~\newcite{hinton2015distilling}, we present a novel method that uses LDA to supervise the training of a deep neural network (DNN), so that the DNN can approximate the costly LDA inference with less computation. Our experiments on a document classification task show that a simple DNN can learn the LDA behavior pretty well, while the inference is speeded up tens or hundreds of times.

cs.LG↗

arXiv2015

VMF-SNE: Embedding for Spherical Data

T-SNE is a well-known approach to embedding high-dimensional data and has been widely used in data visualization. The basic assumption of t-SNE is that the data are non-constrained in the Euclidean space and the local proximity can be modelled by Gaussian distributions. This assumption does not hold for a wide range of data types in practical applications, for instance spherical data for which the local proximity is better modelled by the von Mises-Fisher (vMF) distribution instead of the Gaussian. This paper presents a vMF-SNE embedding algorithm to embed spherical data. An iterative process is derived to produce an efficient embedding. The results on a simulation data set demonstrated that vMF-SNE produces better embeddings than t-SNE for spherical data.

cs.LG↗

arXiv2015

Nanoscale $β$-Nuclear Magnetic Resonance Depth Imaging of Topological Insulators

Considerable evidence suggests that variations in the properties of topological insulators (TIs) at the nanoscale and at interfaces can strongly affect the physics of topological materials. Therefore, a detailed understanding of surface states and interface coupling is crucial to the search for and applications of new topological phases of matter. Currently, no methods can provide depth profiling near surfaces or at interfaces of topologically inequivalent materials. Such a method could advance the study of interactions. Herein we present a non-invasive depth-profiling technique based on $β$-NMR spectroscopy of radioactive $^8$Li$^+$ ions that can provide "one-dimensional imaging" in films of fixed thickness and generates nanoscale views of the electronic wavefunctions and magnetic order at topological surfaces and interfaces. By mapping the $^8$Li nuclear resonance near the surface and 10 nm deep into the bulk of pure and Cr-doped bismuth antimony telluride films, we provide signatures related to the TI properties and their topological non-trivial characteristics that affect the electron-nuclear hyperfine field, the metallic shift and magnetic order. These nanoscale variations in $β$-NMR parameters reflect the unconventional properties of the topological materials under study, and understanding the role of heterogeneities is expected to lead to the discovery of novel phenomena involving quantum materials.

cond-mat.mtrl-sci↗

arXiv2015

Improved Deep Speaker Feature Learning for Text-Dependent Speaker Recognition

A deep learning approach has been proposed recently to derive speaker identifies (d-vector) by a deep neural network (DNN). This approach has been applied to text-dependent speaker recognition tasks and shows reasonable performance gains when combined with the conventional i-vector approach. Although promising, the existing d-vector implementation still can not compete with the i-vector baseline. This paper presents two improvements for the deep learning approach: a phonedependent DNN structure to normalize phone variation, and a new scoring approach based on dynamic time warping (DTW). Experiments on a text-dependent speaker recognition task demonstrated that the proposed methods can provide considerable performance improvement over the existing d-vector implementation.

cs.CL↗

arXiv2015

Recognize Foreign Low-Frequency Words with Similar Pairs

Low-frequency words place a major challenge for automatic speech recognition (ASR). The probabilities of these words, which are often important name entities, are generally under-estimated by the language model (LM) due to their limited occurrences in the training data. Recently, we proposed a word-pair approach to deal with the problem, which borrows information of frequent words to enhance the probabilities of low-frequency words. This paper presents an extension to the word-pair method by involving multiple `predicting words' to produce better estimation for low-frequency words. We also employ this approach to deal with out-of-language words in the task of multi-lingual speech recognition.

cs.CL↗

arXiv2015

Knowledge Transfer Pre-training

Pre-training is crucial for learning deep neural networks. Most of existing pre-training methods train simple models (e.g., restricted Boltzmann machines) and then stack them layer by layer to form the deep structure. This layer-wise pre-training has found strong theoretical foundation and broad empirical support. However, it is not easy to employ such method to pre-train models without a clear multi-layer structure,e.g., recurrent neural networks (RNNs). This paper presents a new pre-training approach based on knowledge transfer learning. In contrast to the layer-wise approach which trains model components incrementally, the new approach trains the entire model as a whole but with an easier objective function. This is achieved by utilizing soft targets produced by a prior trained model (teacher model). Compared to the conventional layer-wise methods, this new method does not care about the model structure, so can be used to pre-train very complex models. Experiments on a speech recognition task demonstrated that with this approach, complex RNNs can be well trained with a weaker deep neural network (DNN) model. Furthermore, the new method can be combined with conventional layer-wise pre-training to deliver additional gains.

cs.LG↗

arXiv2015

Bulk and soft-edge universality for singular values of products of Ginibre random matrices

It has been shown by Akemann, Ipsen and Kieburg that the squared singular values of products of $M$ rectangular random matrices with independent complex Gaussian entries are distributed according to a determinantal point process with a correlation kernel that admits a representation in terms of Meijer G-functions. We prove the universality of the local statistics of the squared singular values, namely, the bulk universality given by the sine kernel and the edge universality given by the Airy kernel. The proof is based on the asymptotic analysis for the double contour integral representation of the correlation kernel. Our strategy can be generalized to deal with other models of products of random matrices introduced recently and to establish similar universal results. Two more examples are investigated, one is the product of $M$ Ginibre matrices and the inverse of $K$ Ginibre matrices studied by Forrester, and the other one is the product of $M-1$ Ginibre matrices with one truncated unitary matrix considered by Kuijlaars and Stivigny.

math.PR↗

arXiv2015

Learning Speech Rate in Speech Recognition

A significant performance reduction is often observed in speech recognition when the rate of speech (ROS) is too low or too high. Most of present approaches to addressing the ROS variation focus on the change of speech signals in dynamic properties caused by ROS, and accordingly modify the dynamic model, e.g., the transition probabilities of the hidden Markov model (HMM). However, an abnormal ROS changes not only the dynamic but also the static property of speech signals, and thus can not be compensated for purely by modifying the dynamic model. This paper proposes an ROS learning approach based on deep neural networks (DNN), which involves an ROS feature as the input of the DNN model and so the spectrum distortion caused by ROS can be learned and compensated for. The experimental results show that this approach can deliver better performance for too slow and too fast utterances, demonstrating our conjecture that ROS impacts both the dynamic and the static property of speech. In addition, the proposed approach can be combined with the conventional HMM transition adaptation method, offering additional performance gains.

cs.CL↗

arXiv2015

Unsupervised Feature Learning with C-SVDDNet

In this paper, we investigate the problem of learning feature representation from unlabeled data using a single-layer K-means network. A K-means network maps the input data into a feature representation by finding the nearest centroid for each input point, which has attracted researchers' great attention recently due to its simplicity, effectiveness, and scalability. However, one drawback of this feature mapping is that it tends to be unreliable when the training data contains noise. To address this issue, we propose a SVDD based feature learning algorithm that describes the density and distribution of each cluster from K-means with an SVDD ball for more robust feature representation. For this purpose, we present a new SVDD algorithm called C-SVDD that centers the SVDD ball towards the mode of local density of each cluster, and we show that the objective of C-SVDD can be solved very efficiently as a linear programming problem. Additionally, traditional unsupervised feature learning methods usually take an average or sum of local representations to obtain global representation which ignore spatial relationship among them. To use spatial information we propose a global representation with a variant of SIFT descriptor. The architecture is also extended with multiple receptive field scales and multiple pooling sizes. Extensive experiments on several popular object recognition benchmarks, such as STL-10, MINST, Holiday and Copydays shows that the proposed C-SVDDNet method yields comparable or better performance than that of the previous state of the art methods.

cs.CV↗

arXiv2015

Deep Speaker Vectors for Semi Text-independent Speaker Verification

Recent research shows that deep neural networks (DNNs) can be used to extract deep speaker vectors (d-vectors) that preserve speaker characteristics and can be used in speaker verification. This new method has been tested on text-dependent speaker verification tasks, and improvement was reported when combined with the conventional i-vector method. This paper extends the d-vector approach to semi text-independent speaker verification tasks, i.e., the text of the speech is in a limited set of short phrases. We explore various settings of the DNN structure used for d-vector extraction, and present a phone-dependent training which employs the posterior features obtained from an ASR system. The experimental results show that it is possible to apply d-vectors on semi text-independent speaker recognition, and the phone-dependent training improves system performance.

cs.CL↗

arXiv2015

Generalized Remote Preparation of Arbitrary $m$-qubit Entangled States via Genuine Entanglements

Herein, we present a feasible, general protocol for quantum communication within a network via generalized remote preparation of an arbitrary $m$-qubit entangled state designed with genuine tripartite Greenberger--Horne--Zeilinger-type entangled resources. During the implementations, we construct novel collective unitary operations; these operations are tasked with performing the necessary phase transfers during remote state preparations. We have distilled our implementation methods into a five-step procedure, which can be used to faithfully recover the desired state during transfer. Compared to previous existing schemes, our methodology features a greatly increased success probability. After the consumption of auxiliary qubits and the performance of collective unitary operations, the probability of successful state transfer is increased four-fold and eight-fold for arbitrary two- and three-qubit entanglements when compared to other methods within the literature, respectively. We conclude this paper with a discussion of the presented scheme for state preparation, including: success probabilities, reducibility and generalizability.

quant-ph↗

arXiv2015

Fermionic non-Abelian fractional Chern insulators from dipolar interactions

We study fermions on a triangular lattice model that exhibits topological flatbands characterized by nonzero Chern numbers. Our scheme stems from the well-known Hofstadter model but the next-nearest-neighbor hopping is introduced, which is crucial for tuning the lowest band to be nearly flat. Differing from previous proposals with the necessity of multiparticle interactions, we consider the more realistic long-range dipolar interaction combined with two-body short-range attractions between fermions. We show the realization of the non-Abelian $ν=1/2$ Moore-Read fractional Chern insulators, and strong evidence for the existence of the more exotic $ν=3/5$ Read-Rezayi fractional Chern insulators. Our results provide insights for the experimental realization of these exotic states by realistic two-body interactions and thus facilitates the implementation of the universal topological quantum computation.

cond-mat.str-el↗

arXiv2015

Demonstration of nonlinear-energy-spread compensation in relativistic electron bunches with corrugated structures

High quality electron beams with flat distributions in both energy and current are critical for many accelerator-based scientific facilities such as free-electron lasers and MeV ultrafast electron diffraction and microscopes. In this Letter we report on using corrugated structures to compensate for the beam nonlinear energy chirp imprinted by the curvature of the radio-frequency field, leading to a significant reduction in beam energy spread. By using a pair of corrugated structures with orthogonal orientations, we show that the quadrupole wake fields which otherwise increase beam emittance can be effectively canceled. This work also extends the applications of corrugated structures to the low beam charge (a few pC) and low beam energy (a few MeV) regime and may have a strong impact in many accelerator-based facilities.

physics.acc-ph↗

arXiv2015

Modularity Enhances the Rate of Evolution in a Rugged Fitness Landscape

Biological systems are modular, and this modularity affects the evolution of biological systems over time and in different environments. We here develop a theory for the dynamics of evolution in a rugged, modular fitness landscape. We show analytically how horizontal gene transfer couples to the modularity in the system and leads to more rapid rates of evolution at short times. The model, in general, analytically demonstrates a selective pressure for the prevalence of modularity in biology. We use this model to show how the evolution of the influenza virus is affected by the modularity of the proteins that are recognized by the human immune system. Approximately 25\% of the observed rate of fitness increase of the virus could be ascribed to a modular viral landscape.

q-bio.PE↗

arXiv2015

Feedback-optimized Extraordinary Optical Transmission of Continuous-variable Entangled States

We report on the feedback-optimized extraordinary optical transmission of continuous-variable entangled states through a hexagonal metal-hole array. The continuous-variable entanglements from a nondegenerate optical parametric amplifier are first demonstrated to survive after a photon-plasmon-photon conversion process. By controlling the reflected light from the metal-hole array, a significant enhancement of quantum correlations has then been experimentally achieved comparing to the case of without such coherent feedback control. This result presents a useful technique to efficiently recover the substantial reflective losses in the plasmonic circuits for quantum information processing.

quant-ph↗