arXiv subjects

Lin Yang

Publications and source records attributed to Lin Yang.

At least 217 records · Page 12Linked to original sources

arXiv2020

Nearly Linear Row Sampling Algorithm for Quantile Regression

We give a row sampling algorithm for the quantile loss function with sample complexity nearly linear in the dimensionality of the data, improving upon the previous best algorithm whose sampling complexity has at least cubic dependence on the dimensionality. Based upon our row sampling algorithm, we give the fastest known algorithm for quantile regression and a graph sparsification algorithm for balanced directed graphs. Our main technical contribution is to show that Lewis weights sampling, which has been used in row sampling algorithms for $\ell_p$ norms, can also be applied in row sampling algorithms for a variety of loss functions. We complement our theoretical results by experiments to demonstrate the practicality of our approach.

cs.DS↗

arXiv2020

Spin Excitations and Spin Wave Gap in the Ferromagnetic Weyl Semimetal Co$_3$Sn$_2$S$_2$

We report a comprehensive neutron scattering study on the spin excitations in the magnetic Weyl semimetal Co$_3$Sn$_2$S$_2$ with quasi-two-dimensional structure. Both in-plane and out-of-plane dispersions of the spin waves are revealed in the ferromagnetic state, similarly dispersive but damped spin excitations persist into the paramagnetic state. The effective exchange interactions have been estimated by a semi-classical Heisenberg model to consistently reproduce the experimental $T_C$ and spin stiffness. However, a full spin wave gap below $E_g=2.3$ meV is observed at $T=4$ K, much larger than the estimated magnetic anisotropy energy ($\sim0.6$ meV), while its temperature dependence indicates a significant contribution from the Weyl fermions. These results suggest that Co$_3$Sn$_2$S$_2$ is a three-dimensional correlated system with large spin stiffness, and the low-energy spin dynamics could interplay with the topological electron states.

cond-mat.str-el↗

arXiv2020

Acoustic Word Embedding System for Code-Switching Query-by-example Spoken Term Detection

In this paper, we propose a deep convolutional neural network-based acoustic word embedding system on code-switching query by example spoken term detection. Different from previous configurations, we combine audio data in two languages for training instead of only using one single language. We transform the acoustic features of keyword templates and searching content to fixed-dimensional vectors and calculate the distances between keyword segments and searching content segments obtained in a sliding manner. An auxiliary variability-invariant loss is also applied to training data within the same word but different speakers. This strategy is used to prevent the extractor from encoding undesired speaker- or accent-related information into the acoustic word embeddings. Experimental results show that our proposed system produces promising searching results in the code-switching test scenario. With the increased number of templates and the employment of variability-invariant loss, the searching performance is further enhanced.

eess.AS↗

arXiv2020

DIHARD II is Still Hard: Experimental Results and Discussions from the DKU-LENOVO Team

In this paper, we present the submitted system for the second DIHARD Speech Diarization Challenge from the DKULENOVO team. Our diarization system includes multiple modules, namely voice activity detection (VAD), segmentation, speaker embedding extraction, similarity scoring, clustering, resegmentation and overlap detection. For each module, we explore different techniques to enhance performance. Our final submission employs the ResNet-LSTM based VAD, the Deep ResNet based speaker embedding, the LSTM based similarity scoring and spectral clustering. Variational Bayes (VB) diarization is applied in the resegmentation stage and overlap detection also brings slight improvement. Our proposed system achieves 18.84% DER in Track1 and 27.90% DER in Track2. Although our systems have reduced the DERs by 27.5% and 31.7% relatively against the official baselines, we believe that the diarization task is still very difficult.

eess.AS↗

arXiv2020

IROS 2019 Lifelong Robotic Vision Challenge -- Lifelong Object Recognition Report

This report summarizes IROS 2019-Lifelong Robotic Vision Competition (Lifelong Object Recognition Challenge) with methods and results from the top $8$ finalists (out of over~$150$ teams). The competition dataset (L)ifel(O)ng (R)obotic V(IS)ion (OpenLORIS) - Object Recognition (OpenLORIS-object) is designed for driving lifelong/continual learning research and application in robotic vision domain, with everyday objects in home, office, campus, and mall scenarios. The dataset explicitly quantifies the variants of illumination, object occlusion, object size, camera-object distance/angles, and clutter information. Rules are designed to quantify the learning capability of the robotic vision system when faced with the objects appearing in the dynamic environments in the contest. Individual reports, dataset information, rules, and released source code can be found at the project homepage: "https://lifelong-robotic-vision.github.io/competition/".

cs.CV↗

arXiv2020

Multi-modal Sentiment Analysis using Super Characters Method on Low-power CNN Accelerator Device

Recent years NLP research has witnessed the record-breaking accuracy improvement by DNN models. However, power consumption is one of the practical concerns for deploying NLP systems. Most of the current state-of-the-art algorithms are implemented on GPUs, which is not power-efficient and the deployment cost is also very high. On the other hand, CNN Domain Specific Accelerator (CNN-DSA) has been in mass production providing low-power and low cost computation power. In this paper, we will implement the Super Characters method on the CNN-DSA. In addition, we modify the Super Characters method to utilize the multi-modal data, i.e. text plus tabular data in the CL-Aff sharedtask.

cs.CL↗

arXiv2019

End-to-End Adversarial Shape Learning for Abdomen Organ Deep Segmentation

Automatic segmentation of abdomen organs using medical imaging has many potential applications in clinical workflows. Recently, the state-of-the-art performance for organ segmentation has been achieved by deep learning models, i.e., convolutional neural network (CNN). However, it is challenging to train the conventional CNN-based segmentation models that aware of the shape and topology of organs. In this work, we tackle this problem by introducing a novel end-to-end shape learning architecture -- organ point-network. It takes deep learning features as inputs and generates organ shape representations as points that located on organ surface. We later present a novel adversarial shape learning objective function to optimize the point-network to capture shape information better. We train the point-network together with a CNN-based segmentation model in a multi-task fashion so that the shared network parameters can benefit from both shape learning and segmentation tasks. We demonstrate our method with three challenging abdomen organs including liver, spleen, and pancreas. The point-network generates surface points with fine-grained details and it is found critical for improving organ segmentation. Consequently, the deep segmentation model is improved by the introduced shape learning as significantly better Dice scores are observed for spleen and pancreas segmentation.

cs.CV↗

arXiv2019

Correlating the nanoscale structural, magnetic and magneto-transport properties in SrRuO3-based perovskite oxide ultra-thin films

We investigated the structural and magnetic properties of bare SrRuO$_3$ (SRO) ultra-thin films and SrRuO$_3$/SrIrO$_3$/SrZrO$_3$ (SRO/SIO/SZO: RIZ) trilayer heterostructures between 10 K and 80 K, by comparing macroscopic data using magneto-optical Kerr effect (MOKE) and magneto-transport (anomalous Hall effect: AHE), with nanoscale fingerprints when applying non-contact scanning force microscopy (nc-SFM) and magnetic force microscopy (MFM). SRO and RIZ ultra-thin films were epitaxially grown at 650C onto vicinal SrTiO$_3$ (100) single-crystalline substrates to a nominal thickness of 4 and 4/2/2 unit cells (uc), respectively. Our correlated analysis allows associating topographic sample features of overgrown individual layers to their residual magnetization, as is shown here to be relevant for interpreting the macroscopic AHE data. Although the hump-like features in the AHE suggest a magnetically extured skyrmion phase to exist around 55 K associated to the topological Hall effect (THE), both our MOKE and MFM data cannot support this theory. In contrast, our SFM/MFM local-scale analysis finds the local coercive field to be strongly dependent on the effective layer thickness and stoichiometry in both the SRO and RIZ samples, with huge impact on the local band-structure. In fact, it is these variations that in turn mimic a potential THE through anomalies in the AHE resistivity loops.

physics.app-ph↗

arXiv2019

Squared English Word: A Method of Generating Glyph to Use Super Characters for Sentiment Analysis

The Super Characters method addresses sentiment analysis problems by first converting the input text into images and then applying 2D-CNN models to classify the sentiment. It achieves state of the art performance on many benchmark datasets. However, it is not as straightforward to apply in Latin languages as in Asian languages. Because the 2D-CNN model is designed to recognize two-dimensional images, it is better if the inputs are in the form of glyphs. In this paper, we propose SEW (Squared English Word) method generating a squared glyph for each English word by drawing Super Characters images of each English word at the alphabet level, combining the squared glyph together into a whole Super Characters image at the sentence level, and then applying the CNN model to classify the sentiment within the sentence. We applied the SEW method to Wikipedia dataset and obtained a 2.1% accuracy gain compared to the original Super Characters method. For multi-modal data with both structured tabular data and unstructured natural language text, the modified SEW method integrates the data into a single image and classifies sentiment with one unified CNN model.

cs.CL↗

arXiv2019

SuperTML: Two-Dimensional Word Embedding for the Precognition on Structured Tabular Data

Tabular data is the most commonly used form of data in industry. Gradient Boosting Trees, Support Vector Machine, Random Forest, and Logistic Regression are typically used for classification tasks on tabular data. DNN models using categorical embeddings are also applied in this task, but all attempts thus far have used one-dimensional embeddings. The recent work of Super Characters method using two-dimensional word embeddings achieved the state of art result in text classification tasks, showcasing the promise of this new approach. In this paper, we propose the SuperTML method, which borrows the idea of Super Characters method and two-dimensional embeddings to address the problem of classification on tabular data. For each input of tabular data, the features are first projected into two-dimensional embeddings like an image, and then this image is fed into fine-tuned two-dimensional CNN models for classification. Experimental results have shown that the proposed SuperTML method had achieved state-of-the-art results on both large and small datasets.

cs.CV↗

arXiv2019

SuperCaptioning: Image Captioning Using Two-dimensional Word Embedding

Language and vision are processed as two different modal in current work for image captioning. However, recent work on Super Characters method shows the effectiveness of two-dimensional word embedding, which converts text classification problem into image classification problem. In this paper, we propose the SuperCaptioning method, which borrows the idea of two-dimensional word embedding from Super Characters method, and processes the information of language and vision together in one single CNN model. The experimental results on Flickr30k data shows the proposed method gives high quality image captions. An interactive demo is ready to show at the workshop.

cs.CL↗

arXiv2019

System Demo for Transfer Learning across Vision and Text using Domain Specific CNN Accelerator for On-Device NLP Applications

Power-efficient CNN Domain Specific Accelerator (CNN-DSA) chips are currently available for wide use in mobile devices. These chips are mainly used in computer vision applications. However, the recent work of Super Characters method for text classification and sentiment analysis tasks using two-dimensional CNN models has also achieved state-of-the-art results through the method of transfer learning from vision to text. In this paper, we implemented the text classification and sentiment analysis applications on mobile devices using CNN-DSA chips. Compact network representations using one-bit and three-bits precision for coefficients and five-bits for activations are used in the CNN-DSA chip with power consumption less than 300mW. For edge devices under memory and compute constraints, the network is further compressed by approximating the external Fully Connected (FC) layers within the CNN-DSA chip. At the workshop, we have two system demonstrations for NLP tasks. The first demo classifies the input English Wikipedia sentence into one of the 14 ontologies. The second demo classifies the Chinese online-shopping review into positive or negative.

cs.CL↗

arXiv2019

SuperChat: Dialogue Generation by Transfer Learning from Vision to Language using Two-dimensional Word Embedding and Pretrained ImageNet CNN Models

The recent work of Super Characters method using two-dimensional word embedding achieved state-of-the-art results in text classification tasks, showcasing the promise of this new approach. This paper borrows the idea of Super Characters method and two-dimensional embedding, and proposes a method of generating conversational response for open domain dialogues. The experimental results on a public dataset shows that the proposed SuperChat method generates high quality responses. An interactive demo is ready to show at the workshop.

cs.CL↗

arXiv2019

2-bit Model Compression of Deep Convolutional Neural Network on ASIC Engine for Image Retrieval

Image retrieval utilizes image descriptors to retrieve the most similar images to a given query image. Convolutional neural network (CNN) is becoming the dominant approach to extract image descriptors for image retrieval. For low-power hardware implementation of image retrieval, the drawback of CNN-based feature descriptor is that it requires hundreds of megabytes of storage. To address this problem, this paper applies deep model quantization and compression to CNN in ASIC chip for image retrieval. It is demonstrated that the CNN-based features descriptor can be extracted using as few as 2-bit weights quantization to deliver a similar performance as floating-point model for image retrieval. In addition, to implement CNN in ASIC, especially for large scale images, the limited buffer size of chips should be considered. To retrieve large scale images, we propose an improved pooling strategy, region nested invariance pooling (RNIP), which uses cropped sub-images for CNN. Testing results on chip show that integrating RNIP with the proposed 2-bit CNN model compression approach is capable of retrieving large scale images.

cs.AI↗

arXiv2019

Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network

Synthesized medical images have several important applications, e.g., as an intermedium in cross-modality image registration and as supplementary training samples to boost the generalization capability of a classifier. Especially, synthesized computed tomography (CT) data can provide X-ray attenuation map for radiation therapy planning. In this work, we propose a generic cross-modality synthesis approach with the following targets: 1) synthesizing realistic looking 3D images using unpaired training data, 2) ensuring consistent anatomical structures, which could be changed by geometric distortion in cross-modality synthesis and 3) improving volume segmentation by using synthetic data for modalities with limited training samples. We show that these goals can be achieved with an end-to-end 3D convolutional neural network (CNN) composed of mutually-beneficial generators and segmentors for image synthesis and segmentation tasks. The generators are trained with an adversarial loss, a cycle-consistency loss, and also a shape-consistency loss, which is supervised by segmentors, to reduce the geometric distortion. From the segmentation view, the segmentors are boosted by synthetic data from generators in an online manner. Generators and segmentors prompt each other alternatively in an end-to-end training fashion. With extensive experiments on a dataset including a total of 4,496 CT and magnetic resonance imaging (MRI) cardiovascular volumes, we show both tasks are beneficial to each other and coupling these two tasks results in better performance than solving them exclusively.

cs.CV↗

arXiv2019

The direction of the $d$-vector in a nematic triplet superconductor

We investigate the states of triplet pairing in a candidate nematic superconductor versus typical material parameters, using the mean field theory for two- and three-dimensional tight-binding models with local triplet pairing in the $E_u$ representation. In the two-dimensional model, the system favors the fully gapped chiral state for weaker warping or lower filling level, while a nodal and nematic $Δ_{4x}$ state is favorable for stronger warping or higher filling, with the $d$-vector aligned along the principle axis. In the presence of lattice distortion, relative elongation along one of the principle axes, ${\bf a}$, tends to rotate the nematic $d$-vector orthogonal to ${\bf a}$, resulting in the nematic $Δ_{4y}$ state at sufficient elongation. Three-dimensionality is seen to suppress the chiral state in favor of the nematic ones. Our results may explain the variety in the probed direction of the $d$-vector in existing experiments.

cond-mat.supr-con↗

arXiv2019

SPDA: Superpixel-based Data Augmentation for Biomedical Image Segmentation

Supervised training a deep neural network aims to "teach" the network to mimic human visual perception that is represented by image-and-label pairs in the training data. Superpixelized (SP) images are visually perceivable to humans, but a conventionally trained deep learning model often performs poorly when working on SP images. To better mimic human visual perception, we think it is desirable for the deep learning model to be able to perceive not only raw images but also SP images. In this paper, we propose a new superpixel-based data augmentation (SPDA) method for training deep learning models for biomedical image segmentation. Our method applies a superpixel generation scheme to all the original training images to generate superpixelized images. The SP images thus obtained are then jointly used with the original training images to train a deep learning model. Our experiments of SPDA on four biomedical image datasets show that SPDA is effective and can consistently improve the performance of state-of-the-art fully convolutional networks for biomedical image segmentation in 2D and 3D images. Additional studies also demonstrate that SPDA can practically reduce the generalization gap.

cs.CV↗

arXiv2019

Effects of dopants on the glass forming ability in Al-based metallic alloy

The effect of dopants on the metallic glass forming ability is usually considered based on analysis of changes in the liquid structure or thermodynamics. What is missing in such considerations is an analysis of how a dopant changes the properties of the crystal phases which can form instead of the glass. In order to illuminate this aspect we performed molecular dynamics simulations to study the effects of Mg and Sm dopants on the crystal nucleation in Al. The simulation data were found to be consistent with the experimental observations that addition of Mg to Al does not lead to vitrification but addition of only 8% Sm does. The significant effect of Sm doping was related to the intolerance of Al to this dopant. This leads to increase in the solid-liquid interfacial free energy, and therefore, to increase in the nucleation barrier and to dramatic decrease in the nucleation rate. The intolerance mechanism also significantly affects the growth kinetics.

cond-mat.mtrl-sci↗