arXiv subjects

Yi Yang

Publications and source records attributed to Yi Yang.

At least 1,045 records · Page 58Linked to original sources

arXiv2017

Random Erasing Data Augmentation

In this paper, we introduce Random Erasing, a new data augmentation method for training the convolutional neural network (CNN). In training, Random Erasing randomly selects a rectangle region in an image and erases its pixels with random values. In this process, training images with various levels of occlusion are generated, which reduces the risk of over-fitting and makes the model robust to occlusion. Random Erasing is parameter learning free, easy to implement, and can be integrated with most of the CNN-based recognition models. Albeit simple, Random Erasing is complementary to commonly used data augmentation techniques such as random cropping and flipping, and yields consistent improvement over strong baselines in image classification, object detection and person re-identification. Code is available at: https://github.com/zhunzhong07/Random-Erasing.

cs.CV↗

arXiv2017

String Scattering Amplitudes and Deformed Cubic String Field Theory

We study string scattering amplitudes by using the deformed cubic string field theory which is equivalent to the string field theory in the proper-time gauge. The four-string scattering amplitudes with three tachyons and an arbitrary string state are calculated. The string field theory yields the string scattering amplitudes evaluated on the world sheet of string scattering whereas the coventional method, based on the first quantized theory brings us the string scattering amplitudes defined on the upper half plane. For the highest spin states, generated by the primary operators, both calculations are in perfect agreement. In this case, the string scattering amplitudes are invariant under the conformal transformation, which maps the string world sheet onto the upper half plane. If the external string states are general massive states, generated by non-primary field operators, we need to take into account carefully the conformal transformation between the world sheet and the upper half plane. We show by an explicit calculation that the string scattering amplitudes calculated by using the deformed cubic string field theory transform into those of the first quantized theory on the upper half plane by the conformal transformation, generated by the Schwarz-Christoffel mapping.

hep-th↗

arXiv2017

Tunable magnetization relaxation of Fe_{2}Cr_{1-x}Co_{x}Si half-metallic Heusler alloys by band structure engineering

We report a systematic investigation on the magnetization relaxation properties of iron-based half-metallic Heusler alloy Fe$_{2}$Cr$_{1-x}$Co_${x}$Si (FCCS) thin films using broadband angular-resolved ferromagnetic resonance. Band structure engineering through Co doping (x) demonstrated by first-principles calculations is shown to tune the intrinsic magnetic damping over an order of magnitude, namely 0.01-0.0008. Notably, the intrinsic damping constants for samples with high Co concentration are among the lowest reported for Heusler alloys and even comparable to magnetic insulator yttrium iron garnet. Furthermore, a significant reduction of both isotropic and anisotropic contributions of extrinsic damping of the FCCS alloys was found in the FCCS films with x=0.5-0.75, which is of particular importance for applications. These results demonstrate a practical recipe to tailor functional magnetization for Heusler alloy-based spintronics at room temperature

cond-mat.mtrl-sci↗

arXiv2017

LPG: a four-groups probabilistic approach to leveraging pleiotropy in genome-wide association studies

To date, genome-wide association studies (GWAS) have successfully identified tens of thousands of genetic variants among a variety of traits/diseases, shedding a light on the genetic architecture of complex diseases. Polygenicity of complex diseases, which refers to the phenomenon that a vast number of risk variants collectively contribute to the heritability of complex diseases with modest individual effects, have been widely accepted. This imposes a major challenge towards fully characterizing the genetic bases of complex diseases. An immediate implication of polygenicity is that a much larger sample size is required to detect risk variants with weak/moderate effects. Meanwhile, accumulating evidence suggests that different complex diseases can share genetic risk variants, a phenomenon known as pleiotropy. In this study, we propose a statistical framework for Leveraging Pleiotropic effects in large-scale GWAS data (LPG). LPG utilizes a variational Bayesian expectation-maximization (VBEM) algorithm, making it computationally efficient and scalable for genome-wide scale analysis. To demon- strate the advantage of LPG over existing methods that do not leverage pleiotropy, we conducted extensive simulation studies and also applied LPG to analyze three au- toimmune disorders (Crohn's disease, rheumatoid arthritis and Type 1 diabetes). The results indicate that LPG can improve the power of prioritization of risk variants and accuracy of risk prediction by leveraging pleiotropy. The software is available at http- s://github.com/Shufeyangyi2015310117/LPG.

stat.ME↗

arXiv2017

Nanophotonic Particle Simulation and Inverse Design Using Artificial Neural Networks

We propose a method to use artificial neural networks to approximate light scattering by multilayer nanoparticles. We find the network needs to be trained on only a small sampling of the data in order to approximate the simulation to high precision. Once the neural network is trained, it can simulate such optical processes orders of magnitude faster than conventional simulations. Furthermore, the trained neural network can be used solve nanophotonic inverse design problems by using back- propogation - where the gradient is analytical, not numerical.

physics.comp-ph↗

arXiv2017

Smith-Purcell Radiation from Low-Energy Electrons

Recent advances in the fabrication of nanostructures and nanoscale features in metasurfaces offer a new prospect for generating visible, light emission from low energy electrons. In this paper, we present the experimental observation of visible light emission from low-energy free electrons interacting with nanoscale periodic surfaces through the Smith-Purcell (SP) effect. SP radiation is emitted when electrons pass in close proximity over a periodic structure, inducing collective charge motion or dipole excitations near the surface, thereby giving rise to electromagnetic radiation. We demonstrate a controlled emission of SP light from nanoscale gold gratings with periodicity as small as 50 nm, enabling the observation of visible SP radiation by low energy electrons (1.5 to 6 keV), an order of magnitude lower than previously reported. We study the emission wavelength and intensity dependence on the grating pitch and electron energy, showing agreement between experiment and theory. Further reduction of structure periodicity should enable the production of SP-based devices that operate with even slower electrons that allow an even smaller footprint and facilitate the investigation of quantum effects for light generation in nanoscale devices. A tunable light source integrated in an electron microscope would enable the development of novel electron-optical correlated spectroscopic techniques, with additional applications ranging from biological imaging to solid-state lighting.

physics.optics↗

arXiv2017

Spectral and spatial shaping of Smith Purcell Radiation

The Smith Purcell effect, observed when an electron beam passes in the vicinity of a periodic structure, is a promising platform for the generation of electromagnetic radiation in previously-unreachable spectral ranges. However, most of the studies of this radiation were performed on simple periodic gratings, whose radiation spectrum exhibits a single peak and its higher harmonics predicted by a well-established dispersion relation. Here, we propose a method to shape the spatial and spectral far-field distribution of the radiation using complex periodic and aperiodic gratings. We show, theoretically and experimentally, that engineering multiple peak spectra with controlled widths located at desired wavelengths is achievable using Smith-Purcell radiation. Our method opens the way to free-electron driven sources with tailored angular and spectral response, and gives rise to focusing functionality for spectral ranges where lenses are unavailable or inefficient.

physics.optics↗

arXiv2017

Dynamic Computational Time for Visual Attention

We propose a dynamic computational time model to accelerate the average processing time for recurrent visual attention (RAM). Rather than attention with a fixed number of steps for each input image, the model learns to decide when to stop on the fly. To achieve this, we add an additional continue/stop action per time step to RAM and use reinforcement learning to learn both the optimal attention policy and stopping policy. The modification is simple but could dramatically save the average computational time while keeping the same recognition performance as RAM. Experimental results on CUB-200-2011 and Stanford Cars dataset demonstrate the dynamic computational model can work effectively for fine-grained image recognition.The source code of this paper can be obtained from https://github.com/baidu-research/DT-RAM

cs.CV↗

arXiv2017

Robust PCA by Manifold Optimization

Robust PCA is a widely used statistical procedure to recover a underlying low-rank matrix with grossly corrupted observations. This work considers the problem of robust PCA as a nonconvex optimization problem on the manifold of low-rank matrices, and proposes two algorithms (for two versions of retractions) based on manifold optimization. It is shown that, with a proper designed initialization, the proposed algorithms are guaranteed to converge to the underlying low-rank matrix linearly. Compared with a previous work based on the Burer-Monterio decomposition of low-rank matrices, the proposed algorithms reduce the dependence on the conditional number of the underlying low-rank matrix theoretically. Simulations and real data examples confirm the competitive performance of our method.

stat.ML↗

arXiv2017

Overcoming Language Variation in Sentiment Analysis with Social Attention

Variation in language is ubiquitous, particularly in newer forms of writing such as social media. Fortunately, variation is not random, it is often linked to social properties of the author. In this paper, we show how to exploit social networks to make sentiment analysis more robust to social language variation. The key idea is linguistic homophily: the tendency of socially linked individuals to use language in similar ways. We formalize this idea in a novel attention-based neural network architecture, in which attention is divided among several basis models, depending on the author's position in the social network. This has the effect of smoothing the classification function across the social network, and makes it possible to induce personalized classifiers even for authors for whom there is no labeled data or demographic metadata. This model significantly improves the accuracies of sentiment analysis on Twitter and on review data.

cs.CL↗

arXiv2017

Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro

The main contribution of this paper is a simple semi-supervised pipeline that only uses the original training set without collecting extra data. It is challenging in 1) how to obtain more training data only from the training set and 2) how to use the newly generated data. In this work, the generative adversarial network (GAN) is used to generate unlabeled samples. We propose the label smoothing regularization for outliers (LSRO). This method assigns a uniform label distribution to the unlabeled images, which regularizes the supervised model and improves the baseline. We verify the proposed method on a practical problem: person re-identification (re-ID). This task aims to retrieve a query person from other cameras. We adopt the deep convolutional generative adversarial network (DCGAN) for sample generation, and a baseline convolutional neural network (CNN) for representation learning. Experiments show that adding the GAN-generated data effectively improves the discriminative ability of learned CNN embeddings. On three large-scale datasets, Market-1501, CUHK03 and DukeMTMC-reID, we obtain +4.37%, +1.6% and +2.46% improvement in rank-1 precision over the baseline CNN, respectively. We additionally apply the proposed method to fine-grained bird recognition and achieve a +0.6% improvement over a strong baseline. The code is available at https://github.com/layumi/Person-reID_GAN.

cs.CV↗

arXiv2017

An Improved Residual LSTM Architecture for Acoustic Modeling

Long Short-Term Memory (LSTM) is the primary recurrent neural networks architecture for acoustic modeling in automatic speech recognition systems. Residual learning is an efficient method to help neural networks converge easier and faster. In this paper, we propose several types of residual LSTM methods for our acoustic modeling. Our experiments indicate that, compared with classic LSTM, our architecture shows more than 8% relative reduction in Phone Error Rate (PER) on TIMIT tasks. At the same time, our residual fast LSTM approach shows 4% relative reduction in PER on the same task. Besides, we find that all this architecture could have good results on THCHS-30, Librispeech and Switchboard corpora.

cs.CL↗

arXiv2017

Solving Lauricella String Scattering Amplitudes through Recurrence Relations

We show that there exist infinite number of recurrence relations valid for all energies among the open bosonic string scattering amplitudes (SSA) of three tachyons and one arbitrary string state, or the Lauricella SSA. Moreover, these infinite number of recurrence relations can be used to solve all the Lauricella SSA and express them in terms of one single four tachyon amplitude. These results extend the solvability of SSA at the high energy, fixed angle scattering limit and those at the Regge scattering limit discovered previously.

hep-th↗

arXiv2017

PatchShuffle Regularization

This paper focuses on regularizing the training of the convolutional neural network (CNN). We propose a new regularization approach named ``PatchShuffle`` that can be adopted in any classification-oriented CNN models. It is easy to implement: in each mini-batch, images or feature maps are randomly chosen to undergo a transformation such that pixels within each local patch are shuffled. Through generating images and feature maps with interior orderless patches, PatchShuffle creates rich local variations, reduces the risk of network overfitting, and can be viewed as a beneficial supplement to various kinds of training regularization techniques, such as weight decay, model ensemble and dropout. Experiments on four representative classification datasets show that PatchShuffle improves the generalization ability of CNN especially when the data is scarce. Moreover, we empirically illustrate that CNN models trained with PatchShuffle are more robust to noise and local changes in an image.

cs.CV↗

arXiv2017

UTS submission to Google YouTube-8M Challenge 2017

In this paper, we present our solution to Google YouTube-8M Video Classification Challenge 2017. We leveraged both video-level and frame-level features in the submission. For video-level classification, we simply used a 200-mixture Mixture of Experts (MoE) layer, which achieves GAP 0.802 on the validation set with a single model. For frame-level classification, we utilized several variants of recurrent neural networks, sequence aggregation with attention mechanism and 1D convolutional models. We achieved GAP 0.8408 on the private testing set with the ensemble model. The source code of our models can be found in \url{https://github.com/ffmpbgrnn/yt8m}.

cs.CV↗

arXiv2017

Pedestrian Alignment Network for Large-scale Person Re-identification

Person re-identification (person re-ID) is mostly viewed as an image retrieval problem. This task aims to search a query person in a large image pool. In practice, person re-ID usually adopts automatic detectors to obtain cropped pedestrian images. However, this process suffers from two types of detector errors: excessive background and part missing. Both errors deteriorate the quality of pedestrian alignment and may compromise pedestrian matching due to the position and scale variances. To address the misalignment problem, we propose that alignment can be learned from an identification procedure. We introduce the pedestrian alignment network (PAN) which allows discriminative embedding learning and pedestrian alignment without extra annotations. Our key observation is that when the convolutional neural network (CNN) learns to discriminate between different identities, the learned feature maps usually exhibit strong activations on the human body rather than the background. The proposed network thus takes advantage of this attention mechanism to adaptively locate and align pedestrians within a bounding box. Visual examples show that pedestrians are better aligned with PAN. Experiments on three large-scale re-ID datasets confirm that PAN improves the discriminative ability of the feature embeddings and yields competitive accuracy with the state-of-the-art methods.

cs.CV↗

arXiv2017

Unsupervised Person Re-identification: Clustering and Fine-tuning

The superiority of deeply learned pedestrian representations has been reported in very recent literature of person re-identification (re-ID). In this paper, we consider the more pragmatic issue of learning a deep feature with no or only a few labels. We propose a progressive unsupervised learning (PUL) method to transfer pretrained deep representations to unseen domains. Our method is easy to implement and can be viewed as an effective baseline for unsupervised re-ID feature learning. Specifically, PUL iterates between 1) pedestrian clustering and 2) fine-tuning of the convolutional neural network (CNN) to improve the original model trained on the irrelevant labeled dataset. Since the clustering results can be very noisy, we add a selection operation between the clustering and fine-tuning. At the beginning when the model is weak, CNN is fine-tuned on a small amount of reliable examples which locate near to cluster centroids in the feature space. As the model becomes stronger in subsequent iterations, more images are being adaptively selected as CNN training samples. Progressively, pedestrian clustering and the CNN model are improved simultaneously until algorithm convergence. This process is naturally formulated as self-paced learning. We then point out promising directions that may lead to further improvement. Extensive experiments on three large-scale re-ID datasets demonstrate that PUL outputs discriminative features that improve the re-ID accuracy.

cs.CV↗

arXiv2017

All-angle negative refraction of highly squeezed plasmon and phonon polaritons in graphene-boron nitride heterostructures

A fundamental building block for nanophotonics is the ability to achieve negative refraction of polaritons, because this could enable the demonstration of many unique nanoscale applications such as deep-subwavelength imaging, superlens, and novel guiding. However, to achieve negative refraction of highly squeezed polaritons, such as plasmon polaritons in graphene and phonon polaritons in boron nitride (BN) with their wavelengths squeezed by a factor over 100, requires the ability to flip the sign of their group velocity at will, which is challenging. Here we reveal that the strong coupling between plasmon and phonon polaritons in graphene-BN heterostructures can be used to flip the sign of the group velocity of the resulting hybrid (plasmon-phonon-polariton) modes. We predict all-angle negative refraction between plasmon and phonon polaritons, and even more surprisingly, between hybrid graphene plasmons, and between hybrid phonon polaritons. Graphene-BN heterostructures thus provide a versatile platform for the design of nano-metasurfaces and nano-imaging elements.

physics.optics↗