arXiv subjects

Yong Xu

Publications and source records attributed to Yong Xu.

At least 451 records · Page 25Linked to original sources

arXiv2017

Discriminative Block-Diagonal Representation Learning for Image Recognition

Existing block-diagonal representation researches mainly focuses on casting block-diagonal regularization on training data, while only little attention is dedicated to concurrently learning both block-diagonal representations of training and test data. In this paper, we propose a discriminative block-diagonal low-rank representation (BDLRR) method for recognition. In particular, the elaborate BDLRR is formulated as a joint optimization problem of shrinking the unfavorable representation from off-block-diagonal elements and strengthening the compact block-diagonal representation under the semi-supervised framework of low-rank representation. To this end, we first impose penalty constraints on the negative representation to eliminate the correlation between different classes such that the incoherence criterion of the extra-class representation is boosted. Moreover, a constructed subspace model is developed to enhance the self-expressive power of training samples and further build the representation bridge between the training and test samples, such that the coherence of the learned intra-class representation is consistently heightened. Finally, the resulting optimization problem is solved elegantly by employing an alternative optimization strategy, and a simple recognition algorithm on the learned representation is utilized for final prediction. Extensive experimental results demonstrate that the proposed method achieves superb recognition results on four face image datasets, three character datasets, and the fifteen scene multi-categories dataset. It not only shows superior potential on image recognition but also outperforms state-of-the-art methods.

cs.CV↗

arXiv2017

Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation

In domain adaptation, maximum mean discrepancy (MMD) has been widely adopted as a discrepancy metric between the distributions of source and target domains. However, existing MMD-based domain adaptation methods generally ignore the changes of class prior distributions, i.e., class weight bias across domains. This remains an open problem but ubiquitous for domain adaptation, which can be caused by changes in sample selection criteria and application scenarios. We show that MMD cannot account for class weight bias and results in degraded domain adaptation performance. To address this issue, a weighted MMD model is proposed in this paper. Specifically, we introduce class-specific auxiliary weights into the original MMD for exploiting the class prior probability on source and target domains, whose challenge lies in the fact that the class label in target domain is unavailable. To account for it, our proposed weighted MMD model is defined by introducing an auxiliary weight for each class in the source domain, and a classification EM algorithm is suggested by alternating between assigning the pseudo-labels, estimating auxiliary weights and updating model parameters. Extensive experiments demonstrate the superiority of our weighted MMD over conventional MMD for domain adaptation.

cs.CV↗

arXiv2017

Multi-Objective Learning and Mask-Based Post-Processing for Deep Neural Network Based Speech Enhancement

We propose a multi-objective framework to learn both secondary targets not directly related to the intended task of speech enhancement (SE) and the primary target of the clean log-power spectra (LPS) features to be used directly for constructing the enhanced speech signals. In deep neural network (DNN) based SE we introduce an auxiliary structure to learn secondary continuous features, such as mel-frequency cepstral coefficients (MFCCs), and categorical information, such as the ideal binary mask (IBM), and integrate it into the original DNN architecture for joint optimization of all the parameters. This joint estimation scheme imposes additional constraints not available in the direct prediction of LPS, and potentially improves the learning of the primary target. Furthermore, the learned secondary information as a byproduct can be used for other purposes, e.g., the IBM-based post-processing in this work. A series of experiments show that joint LPS and MFCC learning improves the SE performance, and IBM-based post-processing further enhances listening quality of the reconstructed speech.

cs.SD↗

arXiv2017

Attention and Localization based on a Deep Convolutional Recurrent Model for Weakly Supervised Audio Tagging

Audio tagging aims to perform multi-label classification on audio chunks and it is a newly proposed task in the Detection and Classification of Acoustic Scenes and Events 2016 (DCASE 2016) challenge. This task encourages research efforts to better analyze and understand the content of the huge amounts of audio data on the web. The difficulty in audio tagging is that it only has a chunk-level label without a frame-level label. This paper presents a weakly supervised method to not only predict the tags but also indicate the temporal locations of the occurred acoustic events. The attention scheme is found to be effective in identifying the important frames while ignoring the unrelated frames. The proposed framework is a deep convolutional recurrent model with two auxiliary modules: an attention module and a localization module. The proposed algorithm was evaluated on the Task 4 of DCASE 2016 challenge. State-of-the-art performance was achieved on the evaluation set with equal error rate (EER) reduced from 0.13 to 0.11, compared with the convolutional recurrent baseline system.

cs.SD↗

arXiv2017

Convolutional Gated Recurrent Neural Network Incorporating Spatial Features for Audio Tagging

Environmental audio tagging is a newly proposed task to predict the presence or absence of a specific audio event in a chunk. Deep neural network (DNN) based methods have been successfully adopted for predicting the audio tags in the domestic audio scene. In this paper, we propose to use a convolutional neural network (CNN) to extract robust features from mel-filter banks (MFBs), spectrograms or even raw waveforms for audio tagging. Gated recurrent unit (GRU) based recurrent neural networks (RNNs) are then cascaded to model the long-term temporal structure of the audio signal. To complement the input information, an auxiliary CNN is designed to learn on the spatial features of stereo recordings. We evaluate our proposed methods on Task 4 (audio tagging) of the Detection and Classification of Acoustic Scenes and Events 2016 (DCASE 2016) challenge. Compared with our recent DNN-based method, the proposed structure can reduce the equal error rate (EER) from 0.13 to 0.11 on the development set. The spatial features can further reduce the EER to 0.10. The performance of the end-to-end learning on raw waveforms is also comparable. Finally, on the evaluation set, we get the state-of-the-art performance with 0.12 EER while the performance of the best existing system is 0.15 EER.

cs.SD↗

arXiv2016

Manipulating exchange bias using all-optical helicity-dependent switching

Deterministic all-optical control of magnetization without an applied magnetic field has been reported for different materials such as ferrimagnetic and ferromagnetic thin films and granular recording media. These findings have challenged the understanding of all-optical helicity-dependent switching of magnetization and opened many potential applications for future magnetic information, memory and storage technologies. Here we demonstrate optical control of an antiferromagnetic layer through the exchange bias interaction using the helicity of a femtosecond pulsed laser on IrMn/[Co/Pt]xN antiferromagnetic/ ferromagnetic heterostructures. We show controlled switching of the sign of the exchange bias field without any applied field, only by changing the helicity of the light, and quantify the influence of the laser fluence and the number of light pulses on the exchange bias control. We also present the combined effect of laser pulses and applied magnetic field. This study opens applications in spintronic devices where the exchange bias phenomenon is routinely used to fix the magnetization orientation of a magnetic layer in one direction.

cond-mat.mtrl-sci↗

arXiv2016

Experimental observation of a topological band gap opening in ultracold Fermi gases with two-dimensional spin-orbit coupling

The recent experimental realization of synthetic spin-orbit coupling (SOC) opens a new avenue for exploring novel quantum states with ultracold atoms. However, in experiments for generating two-dimensional SOC (e.g., Rashba type), a perpendicular Zeeman field, which opens a band gap at the Dirac point and induces many topological phenomena, is still lacking. Here we theoretically propose and experimentally realize a simple scheme for generating two-dimension SOC and a perpendicular Zeeman field simultaneously in ultracold Fermi gases by tuning the polarization of three Raman lasers that couple three hyperfine ground states of atoms. The resulting band gap opening at the Dirac point is probed using spin injection radio-frequency spectroscopy. Our observation may pave the way for exploring topological transport and topological superfluids with exotic Majorana and Weyl fermion excitations in ultracold atoms.

cond-mat.quant-gas↗

arXiv2016

Unsupervised Feature Learning Based on Deep Models for Environmental Audio Tagging

Environmental audio tagging aims to predict only the presence or absence of certain acoustic events in the interested acoustic scene. In this paper we make contributions to audio tagging in two parts, respectively, acoustic modeling and feature learning. We propose to use a shrinking deep neural network (DNN) framework incorporating unsupervised feature learning to handle the multi-label classification task. For the acoustic modeling, a large set of contextual frames of the chunk are fed into the DNN to perform a multi-label classification for the expected tags, considering that only chunk (or utterance) level rather than frame-level labels are available. Dropout and background noise aware training are also adopted to improve the generalization capability of the DNNs. For the unsupervised feature learning, we propose to use a symmetric or asymmetric deep de-noising auto-encoder (sDAE or aDAE) to generate new data-driven features from the Mel-Filter Banks (MFBs) features. The new features, which are smoothed against background noise and more compact with contextual information, can further improve the performance of the DNN baseline. Compared with the standard Gaussian Mixture Model (GMM) baseline of the DCASE 2016 audio tagging challenge, our proposed method obtains a significant equal error rate (EER) reduction from 0.21 to 0.13 on the development set. The proposed aDAE system can get a relative 6.7% EER reduction compared with the strong DNN baseline on the development set. Finally, the results also show that our approach obtains the state-of-the-art performance with 0.15 EER on the evaluation set of the DCASE 2016 audio tagging task while EER of the first prize of this challenge is 0.17.

cs.SD↗

arXiv2016

Type-II Weyl Points in Three-Dimensional Cold Atom Optical Lattices

Topological Lifshitz phase transition characterizes an abrupt change of the topology of the Fermi surface through a continuous deformation of parameters. Recently, Lifshitz transition has been predicted to separate two types of Weyl points: type-I and type-II (or called structured Weyl points), which has attracted considerable attention in various fields. Although recent experimental investigation has seen a rapid progress on type-II Weyl points, it still remains a significant challenge to observe their characteristic Lifshitz transition. Here, we propose a scheme to realize both type-I and type-II Weyl points in three-dimensional ultracold atomic gases by introducing an experimentally feasible configuration based on current spin-orbit coupling technology. In the resultant Hamiltonian, we find three degenerate points: two Weyl points carrying a Chern number $-1$ and a four-fold degenerate point carrying a Chern number $2$. Remarkably, by continuous tuning of a convenient experimental knob, all these degenerate points can transition from type-I to type-II, thereby providing an ideal platform to study different types of Weyl points and directly probe their Lifshitz phase transition.

cond-mat.quant-gas↗

arXiv2016

Weyl Exceptional Rings in a Three-Dimensional Dissipative Cold Atomic Gas

Three-dimensional topological Weyl semimetals can generally support a zero-dimensional Weyl point characterized by a quantized Chern number or a one-dimensional Weyl nodal ring (or line) characterized by a quantized Berry phase in the momentum space. Here, in a dissipative system with particle gain and loss, we discover a new type of topological ring, dubbed Weyl exceptional ring consisting of exceptional points at which two eigenstates coalesce. Such a Weyl exceptional ring is characterized by both a quantized Chern number and a quantized Berry phase, which are defined via the Riemann surface. We propose an experimental scheme to realize and measure the Weyl exceptional ring in a dissipative cold atomic gas trapped in an optical lattice.

cond-mat.quant-gas↗

arXiv2016

Dynamic Channel Allocation for Interference Mitigation in Relay-assisted Wireless Body Networks

We focus on interference mitigation and energy conservation within a single wireless body area network (WBAN). We adopt two-hop communication scheme supported by the the IEEE 802.15.6 standard (2012). In this paper, we propose a dynamic channel allocation scheme, namely DCAIM to mitigate node-level interference amongst the coexisting regions of a WBAN. At the time, the sensors are in the radius communication of a relay, they form a relay region (RG) coordinated by that relay using time division multiple access (TDMA). In the proposed scheme, each RG creates a table consisting of interfering sensors which it broadcasts to its neighboring sensors. This broadcast allows each pair of RGs to create an interference set (IS). Thus, the members of IS are assigned orthogonal sub-channels whereas other sonsors that do not belong to IS can transmit using the same time slots. Experimental results show that our proposal mitigates node-level interference and improves node and WBAN energy savings. These results are then compared to the results of other schemes. As a result, our scheme outperforms in all cases. Node-level signal to interference and noise ratio (SINR) improved by 11dB whilst, the energy consumption decreased significantly. We further present a probabilistic method and analytically show the outage probability can be effectively reduced to the minimal.

cs.NI↗

arXiv2016

A Joint Detection-Classification Model for Audio Tagging of Weakly Labelled Data

Audio tagging aims to assign one or several tags to an audio clip. Most of the datasets are weakly labelled, which means only the tags of the clip are known, without knowing the occurrence time of the tags. The labeling of an audio clip is often based on the audio events in the clip and no event level label is provided to the user. Previous works have used the bag of frames model assume the tags occur all the time, which is not the case in practice. We propose a joint detection-classification (JDC) model to detect and classify the audio clip simultaneously. The JDC model has the ability to attend to informative and ignore uninformative sounds. Then only informative regions are used for classification. Experimental results on the "CHiME Home" dataset show that the JDC model reduces the equal error rate (EER) from 19.0% to 16.9%. More interestingly, the audio event detector is trained successfully without needing the event level label.

cs.SD↗

arXiv2016

Fully DNN-based Multi-label regression for audio tagging

Acoustic event detection for content analysis in most cases relies on lots of labeled data. However, manually annotating data is a time-consuming task, which thus makes few annotated resources available so far. Unlike audio event detection, automatic audio tagging, a multi-label acoustic event classification task, only relies on weakly labeled data. This is highly desirable to some practical applications using audio analysis. In this paper we propose to use a fully deep neural network (DNN) framework to handle the multi-label classification task in a regression way. Considering that only chunk-level rather than frame-level labels are available, the whole or almost whole frames of the chunk were fed into the DNN to perform a multi-label regression for the expected tags. The fully DNN, which is regarded as an encoding function, can well map the audio features sequence to a multi-tag vector. A deep pyramid structure was also designed to extract more robust high-level features related to the target tags. Further improved methods were adopted, such as the Dropout and background noise aware training, to enhance its generalization capability for new audio recordings in mismatched environments. Compared with the conventional Gaussian Mixture Model (GMM) and support vector machine (SVM) methods, the proposed fully DNN-based method could well utilize the long-term temporal information with the whole chunk as the input. The results show that our approach obtained a 15% relative improvement compared with the official GMM-based method of DCASE 2016 challenge.

cs.CV↗

arXiv2016

Hierarchical learning for DNN-based acoustic scene classification

In this paper, we present a deep neural network (DNN)-based acoustic scene classification framework. Two hierarchical learning methods are proposed to improve the DNN baseline performance by incorporating the hierarchical taxonomy information of environmental sounds. Firstly, the parameters of the DNN are initialized by the proposed hierarchical pre-training. Multi-level objective function is then adopted to add more constraint on the cross-entropy based loss function. A series of experiments were conducted on the Task1 of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2016 challenge. The final DNN-based system achieved a 22.9% relative improvement on average scene classification error as compared with the Gaussian Mixture Model (GMM)-based benchmark system across four standard folds.

cs.SD↗

arXiv2016

Phase tunable Josephson junction and spontaneous mass current in a spin-orbit coupled Fermi superfluid

Atomtronics has the potential for engineering new types of functional devices, such as Josephson junctions (JJs). Previous studies have mainly focused on JJs whose ground states have 0 or $π$ superconducting phase difference across the junctions, while arbitrarily tunable phase JJs may have important applications in superconducting electronics and quantum computation. Here we show that a phase tunable JJ can be implemented in a spin-orbit coupled cold atomic gas with the magnetic tunneling barrier generated by a spin-dependent focused laser beam. We consider the JJ confined in either a linear harmonic trap or a circular ring trap. In the ring trap, the magnetic barrier induces a spontaneous mass current for the ground state of the JJ, demonstrating the magnetoelectric effects of cold atoms.

cond-mat.quant-gas↗

arXiv2016

Thermoelectric Effects and Topological Insulators

The recent discovery of topological insulator (TI) offers new opportunities for the development of thermoelectrics, because many TIs (like Bi$_2$Te$_3$) are excellent thermoelectric (TE) materials. In this review, we will first describe the general TE properties of TIs and show that the coexistence of the bulk and boundary states in TIs introduces unusual TE properties, including strong size effects and anomalous Seebeck effect. Importantly, the TE figure of merit $zT$ of TIs is no longer an intrinsic property, but depends strongly on the geometric size. The geometric parameters of two-dimensional TIs can be tuned to enhance $zT$ to be significantly greater than 1. Then a few proof-of-principle experiments on three-dimensional TIs will be discussed, which observed unconventional TE phenomena that are closely related to the topological nature of the materials. However, current experiments indicate that the metallic surface states, if their advantage of high mobility is not fully utilized, would be detrimental to TE performance. Finally we provide an outlook for future work on topological materials, which offers great possibilities to discover exotic TE effects and may lead to significant breakthroughs in improving $zT$.

cond-mat.mtrl-sci↗

arXiv2016

Natural Scene Character Recognition Using Robust PCA and Sparse Representation

Natural scene character recognition is challenging due to the cluttered background, which is hard to separate from text. In this paper, we propose a novel method for robust scene character recognition. Specifically, we first use robust principal component analysis (PCA) to denoise character image by recovering the missing low-rank component and filtering out the sparse noise term, and then use a simple Histogram of oriented Gradient (HOG) to perform image feature extraction, and finally, use a sparse representation based classifier for recognition. In experiments on four public datasets, namely the Char74K dataset, ICADAR 2003 robust reading dataset, Street View Text (SVT) dataset and IIIT5K-word dataset, our method was demonstrated to be competitive with the state-of-the-art methods.

cs.CV↗

arXiv2016

Structured Weyl Points in Spin-Orbit Coupled Fermionic Superfluids

We demonstrate that a Weyl point, widely examined in 3D Weyl semimetals and superfluids, can develop a pair of non-degenerate gapless spheres. Such a bouquet of two spheres is characterized by three distinct topological invariants of manifolds with full energy gaps, i.e., the Chern number of a 0D point inside one developed sphere, the winding number of a 1D loop around the original Weyl point, and the Chern number of a 2D surface enclosing the whole bouquet. We show that such structured Weyl points can be realized in the superfluid quasiparticle spectrum of a 3D degenerate Fermi gas subject to spin-orbit couplings and Zeeman fields, which supports Fulde-Ferrell superfluids as the ground state.

cond-mat.quant-gas↗