arXiv subjects

Ye Wang

Publications and source records attributed to Ye Wang.

At least 343 records · Page 19Linked to original sources

arXiv2019

Deep Learning-Based Constellation Optimization for Physical Network Coding in Two-Way Relay Networks

This paper studies a new application of deep learning (DL) for optimizing constellations in two-way relaying with physical-layer network coding (PNC), where deep neural network (DNN)-based modulation and demodulation are employed at each terminal and relay node. We train DNNs such that the cross entropy loss is directly minimized, and thus it maximizes the likelihood, rather than considering the Euclidean distance of the constellations. The proposed scheme can be extended to higher level constellations with slight modification of the DNN structure. Simulation results demonstrate a significant performance gain in terms of the achievable sum rate over conventional relaying schemes. Furthermore, since our DNN demodulator directly outputs bit-wise probabilities, it is straightforward to concatenate with soft-decision channel decoding.

cs.IT↗

arXiv2019

Channel Decoding with Quantum Approximate Optimization Algorithm

Motivated by the recent advancement of quantum processors, we investigate quantum approximate optimization algorithm (QAOA) to employ quasi-maximum-likelihood (ML) decoding of classical channel codes. QAOA is a hybrid quantum-classical variational algorithm, which is advantageous for the near-term noisy intermediate-scale quantum (NISQ) devices, where the fidelity of quantum gates is limited by noise and decoherence. We first describe how to construct Ising Hamiltonian model to realize quasi-ML decoding with QAOA. For level-1 QAOA, we derive the systematic way to generate theoretical expressions of cost expectation for arbitrary binary linear codes. Focusing on (7, 4) Hamming code as an example, we analyze the impact of the degree distribution in associated generator matrix on the quantum decoding performance. The excellent performance of higher-level QAOA decoding is verified when Pauli rotation angles are optimized through meta-heuristic variational quantum eigensolver (VQE). Furthermore, we demonstrate the QAOA decoding performance in a real quantum device.

cs.IT↗

arXiv2019

SubSpectralNet - Using Sub-Spectrogram based Convolutional Neural Networks for Acoustic Scene Classification

Acoustic Scene Classification (ASC) is one of the core research problems in the field of Computational Sound Scene Analysis. In this work, we present SubSpectralNet, a novel model which captures discriminative features by incorporating frequency band-level differences to model soundscapes. Using mel-spectrograms, we propose the idea of using band-wise crops of the input time-frequency representations and train a convolutional neural network (CNN) on the same. We also propose a modification in the training method for more efficient learning of the CNN models. We first give a motivation for using sub-spectrograms by giving intuitive and statistical analyses and finally we develop a sub-spectrogram based CNN architecture for ASC. The system is evaluated on the public ASC development dataset provided for the "Detection and Classification of Acoustic Scenes and Events" (DCASE) 2018 Challenge. Our best model achieves an improvement of +14% in terms of classification accuracy with respect to the DCASE 2018 baseline system. Code and figures are available at https://github.com/ssrp/SubSpectralNet

cs.SD↗

arXiv2019

Randomness expansion secured by quantum contextuality

The output randomness from a random number generator can be certified by observing the violation of quantum contextuality inequalities based on the Kochen-Specker theorem. Contextuality can be tested in a single quantum system, which significantly simplifies the experimental requirements to observe the violation comparing to the ones based on nonlocality tests. However, it is not yet resolved how to ensure compatibilities for sequential measurements that is required in contextuality tests. Here, we employ a modified Klyachko-Can-Binicioğlu-Shumovsky contextuality inequality, which can ease the strict compatibility requirement on measurements. On a trapped single \Ba ion system, we experimentally demonstrate violation of the contextuality inequality and realize self-testing quantum random number expansion by closing detection loopholes. We perform $1.29 \times 10^8$ trials of experiments and extract the randomness of $8.06 \times 10^5$ bits with a speed of 270 bits s$^{-1}$. Our demonstration paves the way for the practical high-speed spot-checking quantum random number expansion and other secure information processing applications.

quant-ph↗

arXiv2018

Unsupervised Video Object Segmentation with Distractor-Aware Online Adaptation

Unsupervised video object segmentation is a crucial application in video analysis without knowing any prior information about the objects. It becomes tremendously challenging when multiple objects occur and interact in a given video clip. In this paper, a novel unsupervised video object segmentation approach via distractor-aware online adaptation (DOA) is proposed. DOA models spatial-temporal consistency in video sequences by capturing background dependencies from adjacent frames. Instance proposals are generated by the instance segmentation network for each frame and then selected by motion information as hard negatives if they exist and positives. To adopt high-quality hard negatives, the block matching algorithm is then applied to preceding frames to track the associated hard negatives. General negatives are also introduced in case that there are no hard negatives in the sequence and experiments demonstrate both kinds of negatives (distractors) are complementary. Finally, we conduct DOA using the positive, negative, and hard negative masks to update the foreground/background segmentation. The proposed approach achieves state-of-the-art results on two benchmark datasets, DAVIS 2016 and FBMS-59 datasets.

cs.CV↗

arXiv2018

Towards Visible and Thermal Drone Monitoring with Convolutional Neural Networks

This paper reports a visible and thermal drone monitoring system that integrates deep-learning-based detection and tracking modules. The biggest challenge in adopting deep learning methods for drone detection is the paucity of training drone images especially thermal drone images. To address this issue, we develop two data augmentation techniques. One is a model-based drone augmentation technique that automatically generates visible drone images with a bounding box label on the drone's location. The other is exploiting an adversarial data augmentation methodology to create thermal drone images. To track a small flying drone, we utilize the residual information between consecutive image frames. Finally, we present an integrated detection and tracking system that outperforms the performance of each individual module containing detection or tracking only. The experiments show that even being trained on synthetic data, the proposed system performs well on real-world drone images with complex background. The USC drone detection and tracking dataset with user labeled bounding boxes is available to the public.

cs.CV↗

arXiv2018

Transfer Learning in Brain-Computer Interfaces with Adversarial Variational Autoencoders

We introduce adversarial neural networks for representation learning as a novel approach to transfer learning in brain-computer interfaces (BCIs). The proposed approach aims to learn subject-invariant representations by simultaneously training a conditional variational autoencoder (cVAE) and an adversarial network. We use shallow convolutional architectures to realize the cVAE, and the learned encoder is transferred to extract subject-invariant features from unseen BCI users' data for decoding. We demonstrate a proof-of-concept of our approach based on analyses of electroencephalographic (EEG) data recorded during a motor imagery BCI experiment.

cs.LG↗

arXiv2018

Design Pseudo Ground Truth with Motion Cue for Unsupervised Video Object Segmentation

One major technique debt in video object segmentation is to label the object masks for training instances. As a result, we propose to prepare inexpensive, yet high quality pseudo ground truth corrected with motion cue for video object segmentation training. Our method conducts semantic segmentation using instance segmentation networks and, then, selects the segmented object of interest as the pseudo ground truth based on the motion information. Afterwards, the pseudo ground truth is exploited to finetune the pretrained objectness network to facilitate object segmentation in the remaining frames of the video. We show that the pseudo ground truth could effectively improve the segmentation performance. This straightforward unsupervised video object segmentation method is more efficient than existing methods. Experimental results on DAVIS and FBMS show that the proposed method outperforms state-of-the-art unsupervised segmentation methods on various benchmark datasets. And the category-agnostic pseudo ground truth has great potential to extend to multiple arbitrary object tracking.

cs.CV↗

arXiv2018

An optimized system to solve text-based CAPTCHA

CAPTCHA(Completely Automated Public Turing test to Tell Computers and Humans Apart) can be used to protect data from auto bots. Countless kinds of CAPTCHAs are thus designed, while we most frequently utilize text-based scheme because of most convenience and user-friendly way \cite{bursztein2011text}. Currently, various types of CAPTCHAs need corresponding segmentation to identify single character due to the numerous different segmentation ways. Our goal is to defeat the CAPTCHA, thus firstly the CAPTCHAs need to be split into character by character. There isn't a regular segmentation algorithm to obtain the divided characters in all kinds of examples, which means that we have to treat the segmentation individually. In this paper, we build a whole system to defeat the CAPTCHAs as well as achieve state-of-the-art performance. In detail, we present our self-adaptive algorithm to segment different kinds of characters optimally, and then utilize both the existing methods and our own constructed convolutional neural network as an extra classifier. Results are provided showing how our system work well towards defeating these CAPTCHAs.

cs.CV↗

arXiv2018

Invariant Representations from Adversarially Censored Autoencoders

We combine conditional variational autoencoders (VAE) with adversarial censoring in order to learn invariant representations that are disentangled from nuisance/sensitive variations. In this method, an adversarial network attempts to recover the nuisance variable from the representation, which the VAE is trained to prevent. Conditioning the decoder on the nuisance variable enables clean separation of the representation, since they are recombined for model learning and data reconstruction. We show this natural approach is theoretically well-founded with information-theoretic arguments. Experiments demonstrate that this method achieves invariance while preserving model learning performance, and results in visually improved performance for style transfer and generative sampling tasks.

cs.LG↗

arXiv2018

Frequency stabilization of a 650 nm laser to I$_{2}$ spectrum for trapped $^{138}$Ba$^{+}$ ions

The optical manipulation of Ba$^{+}$ ions is mainly performed by a 493 nm laser for the S$_{1/2}$-P$_{1/2}$ transition and a 650 nm laser for the P$_{1/2}$-D$_{3/2}$ transition. Since the branching ratio between the 493 nm and 650 nm transitions of a single Ba$^{+}$ ion is comparable, stabilization systems of both lasers are equally important for Doppler cooling, sub-Doppler cooling, optical pumping and state detection. The stabilization system of a 493 nm laser to an absolute Te$_2$ reference has been well established. However, the stabilization of a 650 nm laser has not been presented before. Here we report twenty spectral lines of I$_{2}$ in the range of 0.9 GHz above the resonance of the P$_{1/2}$-D$_{3/2}$ transition. We stabilize the 650 nm laser through the optical cavity to the lowest one among these lines, which is about 350 MHz apart, as the absolute frequency reference. Furthermore, we measure the frequency differences between these iodine lines and the Ba$^+$ resonance through fluorescence excitation spectrum with well-resolved dark states, which is in agreement with the theoretical expectation. The presented stabilization scheme enables us to perform precise experiments with Ba$^{+}$ ions.

physics.atom-ph↗

arXiv2018

Bayesian Mosaic: Parallelizable Composite Posterior

This paper proposes Bayesian mosaic, a parallelizable composite posterior, for scalable Bayesian inference on a broad class of multivariate discrete data models. Sampling is embarrassingly parallel since Bayesian mosaic is a multiplication of component posteriors that can be independently sampled from. Analogous to composite likelihood methods, these component posteriors are based on univariate or bivariate marginal densities. Utilizing the fact that the score functions of these densities are unbiased, we show that Bayesian mosaic is consistent and asymptotically normal under mild conditions. Since the evaluation of univariate or bivariate marginal densities can rely on numerical integration, sampling from Bayesian mosaic bypasses the traditional data augmented Markov chain Monte Carlo (DA-MCMC) method, which has a provably slow mixing rate when data are imbalanced. Moreover, we show that sampling from Bayesian mosaic has better scalability to large sample size than DA-MCMC. The method is evaluated via simulation studies and an application on a citation count dataset.

stat.ME↗

arXiv2017

Privacy-Utility Tradeoffs under Constrained Data Release Mechanisms

Privacy-preserving data release mechanisms aim to simultaneously minimize information-leakage with respect to sensitive data and distortion with respect to useful data. Dependencies between sensitive and useful data results in a privacy-utility tradeoff that has strong connections to generalized rate-distortion problems. In this work, we study how the optimal privacy-utility tradeoff region is affected by constraints on the data that is directly available as input to the release mechanism. In particular, we consider the availability of only sensitive data, only useful data, and both (full data). We show that a general hierarchy holds: the tradeoff region given only the sensitive data is no larger than the region given only the useful data, which in turn is clearly no larger than the region given both sensitive and useful data. In addition, we determine conditions under which the tradeoff region given only the useful data coincides with that given full data. These are based on the common information between the sensitive and useful data. We establish these results for general families of privacy and utility measures that satisfy certain natural properties required of any reasonable measure of privacy or utility. We also uncover a new, subtler aspect of the data processing inequality for general non-symmetric privacy measures and discuss its operational relevance and implications. Finally, we derive exact closed-analytic-form expressions for the privacy-utility tradeoffs for symmetrically dependent sensitive and useful data under mutual information and Hamming distortion as the respective privacy and utility measures.

cs.IT↗

arXiv2017

Semantic Segmentation with Reverse Attention

Recent development in fully convolutional neural network enables efficient end-to-end learning of semantic segmentation. Traditionally, the convolutional classifiers are taught to learn the representative semantic features of labeled semantic objects. In this work, we propose a reverse attention network (RAN) architecture that trains the network to capture the opposite concept (i.e., what are not associated with a target class) as well. The RAN is a three-branch network that performs the direct, reverse and reverse-attention learning processes simultaneously. Extensive experiments are conducted to show the effectiveness of the RAN in semantic segmentation. Being built upon the DeepLabv2-LargeFOV, the RAN achieves the state-of-the-art mIoU score (48.1%) for the challenging PASCAL-Context dataset. Significant performance improvements are also observed for the PASCAL-VOC, Person-Part, NYUDv2 and ADE20K datasets.

cs.CV↗

arXiv2017

Single-qubit quantum memory exceeding $10$-minute coherence time

A long-time quantum memory capable of storing and measuring quantum information at the single-qubit level is an essential ingredient for practical quantum computation and com-munication. Recently, there have been remarkable progresses of increasing coherence time for ensemble-based quantum memories of trapped ions, nuclear spins of ionized donors or nuclear spins in a solid. Until now, however, the record of coherence time of a single qubit is on the order of a few tens of seconds demonstrated in trapped ion systems. The qubit coherence time in a trapped ion is mainly limited by the increasing magnetic field fluctuation and the decreasing state-detection efficiency associated with the motional heating of the ion without laser cooling. Here we report the coherence time of a single qubit over $10$ minutes in the hyperfine states of a \Yb ion sympathetically cooled by a \Ba ion in the same Paul trap, which eliminates the heating of the qubit ion even at room temperature. To reach such coherence time, we apply a few thousands of dynamical decoupling pulses to suppress the field fluctuation noise. A long-time quantum memory demonstrated in this experiment makes an important step for construction of the memory zone in scalable quantum computer architectures or for ion-trap-based quantum networks. With further improvement of the coherence time by techniques such as magnetic field shielding and increase of the number of qubits in the quantum memory, our demonstration also makes a basis for other applications including quantum money.

quant-ph↗

arXiv2015

Modeling and analysis of an ultra-stable subluminal laser

We describe a subluminal laser which is extremely stable against perturbations. It makes use of a composite gain spectrum consisting of a broad background along with a narrow peak. The stability of the laser, defined as the change in frequency as a function of a change in the cavity length, is enhanced by a factor given by the group index, which can be as high as 105 for experimentally realizable parameters. We also show that the fundamental linewidth of such a laser is expected to be smaller by the same factor. We first present an analysis where the gain profile is modeled as a superposition of two Lorentzian functions. We then present a numerical study based on a physical scheme for realizing the composite gain profile. In this scheme, the broad gain is produced by a high pressure buffer-gas loaded cell of rubidium vapor. The narrow gain is produced by using a Raman pump in a second rubidium vapor cell, where optical pumping is used to produce a Raman population inversion. We show close agreement between the idealized model and the explicit model. A subluminal laser of this type may prove to be useful for many applications.

physics.optics↗

arXiv2015

Probabilistic Curve Learning: Coulomb Repulsion and the Electrostatic Gaussian Process

Learning of low dimensional structure in multidimensional data is a canonical problem in machine learning. One common approach is to suppose that the observed data are close to a lower-dimensional smooth manifold. There are a rich variety of manifold learning methods available, which allow mapping of data points to the manifold. However, there is a clear lack of probabilistic methods that allow learning of the manifold along with the generative distribution of the observed data. The best attempt is the Gaussian process latent variable model (GP-LVM), but identifiability issues lead to poor performance. We solve these issues by proposing a novel Coulomb repulsive process (Corp) for locations of points on the manifold, inspired by physical models of electrostatic interactions among particles. Combining this process with a GP prior for the mapping function yields a novel electrostatic GP (electroGP) process. Focusing on the simple case of a one-dimensional manifold, we develop efficient inference algorithms, and illustrate substantially improved performance in a variety of experiments including filling in missing frames in video.

stat.ML↗

arXiv2014

An Elementary Completeness Proof for Secure Two-Party Computation Primitives

In the secure two-party computation problem, two parties wish to compute a (possibly randomized) function of their inputs via an interactive protocol, while ensuring that neither party learns more than what can be inferred from only their own input and output. For semi-honest parties and information-theoretic security guarantees, it is well-known that, if only noiseless communication is available, only a limited set of functions can be securely computed; however, if interaction is also allowed over general communication primitives (multi-input/output channels), there are "complete" primitives that enable any function to be securely computed. The general set of complete primitives was characterized recently by Maji, Prabhakaran, and Rosulek leveraging an earlier specialized characterization by Kilian. Our contribution in this paper is a simple, self-contained, alternative derivation using elementary information-theoretic tools.

cs.CR↗