arXiv subjects

Yang Song

Publications and source records attributed to Yang Song.

At least 325 records · Page 18Linked to original sources

arXiv2020

Gaussianization Flows

Iterative Gaussianization is a fixed-point iteration procedure that can transform any continuous random vector into a Gaussian one. Based on iterative Gaussianization, we propose a new type of normalizing flow model that enables both efficient computation of likelihoods and efficient inversion for sample generation. We demonstrate that these models, named Gaussianization flows, are universal approximators for continuous probability distributions under some regularity conditions. Because of this guaranteed expressivity, they can capture multimodal target distributions without compromising the efficiency of sample generation. Experimentally, we show that Gaussianization flows achieve better or comparable performance on several tabular datasets compared to other efficiently invertible flow models such as Real NVP, Glow and FFJORD. In particular, Gaussianization flows are easier to initialize, demonstrate better robustness with respect to different transformations of the training data, and generalize better on small training sets.

cs.LG↗

arXiv2020

Permutation Invariant Graph Generation via Score-Based Generative Modeling

Learning generative models for graph-structured data is challenging because graphs are discrete, combinatorial, and the underlying data distribution is invariant to the ordering of nodes. However, most of the existing generative models for graphs are not invariant to the chosen ordering, which might lead to an undesirable bias in the learned distribution. To address this difficulty, we propose a permutation invariant approach to modeling graphs, using the recent framework of score-based generative modeling. In particular, we design a permutation equivariant, multi-channel graph neural network to model the gradient of the data distribution at the input graph (a.k.a., the score function). This permutation equivariant model of gradients implicitly defines a permutation invariant distribution for graphs. We train this graph neural network with score matching and sample from it with annealed Langevin dynamics. In our experiments, we first demonstrate the capacity of this new architecture in learning discrete graph algorithms. For graph generation, we find that our learning approach achieves better or comparable results to existing models on benchmark datasets.

cs.LG↗

arXiv2020

Toward Predicting Success and Failure in CS2: A Mixed-Method Analysis

Factors driving success and failure in CS1 are the subject of much study but less so for CS2. This paper investigates the transition from CS1 to CS2 in search of leading indicators of success in CS2. Both CS1 and CS2 at the University of North Carolina Wilmington (UNCW) are taught in Python with annual enrollments of 300 and 150 respectively. In this paper, we report on the following research questions: 1) Are CS1 grades indicators of CS2 grades? 2) Does a quantitative relationship exist between CS2 course grade and a modified version of the SCS1 concept inventory? 3) What are the most challenging aspects of CS2, and how well does CS1 prepare students for CS2 from the student's perspective? We provide a quantitative analysis of 2300 CS1 and CS2 course grades from 2013--2019. In Spring 2019, we administered a modified version of the SCS1 concept inventory to 44 students in the first week of CS2. Further, 69 students completed an exit questionnaire at the conclusion of CS2 to gain qualitative student feedback on their challenges in CS2 and on how well CS1 prepared them for CS2. We find that 56% of students' grades were lower in CS2 than CS1, 18% improved their grades, and 26% earned the same grade. Of the changes, 62% were within one grade point. We find a statistically significant correlation between the modified SCS1 score and CS2 grade points. Students identify linked lists and class/object concepts among the most challenging. Student feedback on CS2 challenges and the adequacy of their CS1 preparations identify possible avenues for improving the CS1-CS2 transition.

cs.OH↗

arXiv2020

Learning to Recommend with Multiple Cascading Behaviors

Most existing recommender systems leverage user behavior data of one type only, such as the purchase behavior in E-commerce that is directly related to the business KPI (Key Performance Indicator) of conversion rate. Besides the key behavioral data, we argue that other forms of user behaviors also provide valuable signal, such as views, clicks, adding a product to shop carts and so on. They should be taken into account properly to provide quality recommendation for users. In this work, we contribute a new solution named NMTR (short for Neural Multi-Task Recommendation) for learning recommender systems from user multi-behavior data. We develop a neural network model to capture the complicated and multi-type interactions between users and items. In particular, our model accounts for the cascading relationship among different types of behaviors (e.g., a user must click on a product before purchasing it). To fully exploit the signal in the data of multiple types of behaviors, we perform a joint optimization based on the multi-task learning framework, where the optimization on a behavior is treated as a task. Extensive experiments on two real-world datasets demonstrate that NMTR significantly outperforms state-of-the-art recommender systems that are designed to learn from both single-behavior data and multi-behavior data. Further analysis shows that modeling multiple behaviors is particularly useful for providing recommendation for sparse users that have very few interactions.

cs.IR↗

arXiv2020

Creating and manipulating interfacial spin with giant magnetic response in 4$f$ antiferromagnets

Creating and manipulating spin polarization in low-dimensional electron systems (such as two-dimensional electron gases) is fundamentally essential for spintronic applications, which is yet a challenge to date. In this work, we establish the metamagnetic phase diagram of 4$f$ antiferromagnetic TbScO$_3$ and reveal its giant magnetic response to sub-tesla magnetic field, which has not been reported thus far. Utilizing this giant magnetic response, we demonstrate that the spin polarization of two-dimensional electron gas in SrTiO$_3$/LaTiO$_3$/TbScO$_3$ heterostructure can be manipulated successfully in aid of interfacial 3\textit{d}-4\textit{f} exchange interaction. Remarkably, the hysteretic magnetoresistances of two-dimensional electron gas at the SrTiO$_3$/LaTiO$_3$ interface are entirely determined by the metamagnetic phase transitions of the underlying TbScO$_3$ substrate. Our results pave a novel route to engineer the spin polarization of low-dimensional electron systems in 4$f$ antiferromagnet-based heterostructures.

cond-mat.str-el↗

arXiv2019

Unibody microscope objective tipped with a microsphere: design, fabrication and application in subwavelength imaging

Microsphere-based subwavelength imaging technique was first demonstrated in 2011. After nearly a decade of efforts, such technique has spawned numerous interests in fields such as laser nano-machining, imaging, sensing and biological detection. For wider industrial scale application of the technique, a robust, low-cost objective lens incorporating microsphere lens is highly desired and sought by many researchers. In this work, we demonstrate a unibody microscope objective lens formed by tipping a high-index microsphere onto a Plano-Convex lens and subsequently fitting them into a conventional objective lens. We call such new objective the Plano-Convex-Microsphere (PCM) objective, which resembles the appearance and operation of an ordinary microscope objective, while providing super-resolving power in discerning subwavelength nanoscale features in air condition. The imaging performance of PCM-objective along with the working distance has been systematically investigated. With the assistance of scanning process, larger area imaging is realized. The PCM-objective can be easily adaptive to many existing microscope systems and appealing for commercialization.

physics.optics↗

arXiv2019

Region and Object based Panoptic Image Synthesis through Conditional GANs

Image-to-image translation is significant to many computer vision and machine learning tasks such as image synthesis and video synthesis. It has primary applications in the graphics editing and animation industries. With the development of generative adversarial networks, a lot of attention has been drawn to image-to-image translation tasks. In this paper, we propose and investigate a novel task named as panoptic-level image-to-image translation and a naive baseline of solving this task. Panoptic-level image translation extends the current image translation task to two separate objectives of semantic style translation (adjust the style of objects to that of different domains) and instance transfiguration (swap between different types of objects). The proposed task generates an image from a complete and detailed panoptic perspective which can enrich the context of real-world vision synthesis. Our contribution consists of the proposal of a significant task worth investigating and a naive baseline of solving it. The proposed baseline consists of the multiple instances sequential translation and semantic-level translation with domain-invariant content code.

cs.CV↗

arXiv2019

MintNet: Building Invertible Neural Networks with Masked Convolutions

We propose a new way of constructing invertible neural networks by combining simple building blocks with a novel set of composition rules. This leads to a rich set of invertible architectures, including those similar to ResNets. Inversion is achieved with a locally convergent iterative procedure that is parallelizable and very fast in practice. Additionally, the determinant of the Jacobian can be computed analytically and efficiently, enabling their generative use as flow models. To demonstrate their flexibility, we show that our invertible neural networks are competitive with ResNets on MNIST and CIFAR-10 classification. When trained as generative models, our invertible networks achieve competitive likelihoods on MNIST, CIFAR-10 and ImageNet 32x32, with bits per dimension of 0.98, 3.32 and 4.06 respectively.

cs.LG↗

arXiv2019

Unsupervised Out-of-Distribution Detection with Batch Normalization

Likelihood from a generative model is a natural statistic for detecting out-of-distribution (OoD) samples. However, generative models have been shown to assign higher likelihood to OoD samples compared to ones from the training distribution, preventing simple threshold-based detection rules. We demonstrate that OoD detection fails even when using more sophisticated statistics based on the likelihoods of individual samples. To address these issues, we propose a new method that leverages batch normalization. We argue that batch normalization for generative models challenges the traditional i.i.d. data assumption and changes the corresponding maximum likelihood objective. Based on this insight, we propose to exploit in-batch dependencies for OoD detection. Empirical results suggest that this leads to more robust detection for high-dimensional images.

cs.LG↗

arXiv2019

Representation Learning with Ordered Relation Paths for Knowledge Graph Completion

Incompleteness is a common problem for existing knowledge graphs (KGs), and the completion of KG which aims to predict links between entities is challenging. Most existing KG completion methods only consider the direct relation between nodes and ignore the relation paths which contain useful information for link prediction. Recently, a few methods take relation paths into consideration but pay less attention to the order of relations in paths which is important for reasoning. In addition, these path-based models always ignore nonlinear contributions of path features for link prediction. To solve these problems, we propose a novel KG completion method named OPTransE. Instead of embedding both entities of a relation into the same latent space as in previous methods, we project the head entity and the tail entity of each relation into different spaces to guarantee the order of relations in the path. Meanwhile, we adopt a pooling strategy to extract nonlinear and complex features of different paths to further improve the performance of link prediction. Experimental results on two benchmark datasets show that the proposed model OPTransE performs better than state-of-the-art methods.

cs.LG↗

arXiv2019

Geo-Aware Networks for Fine-Grained Recognition

Fine-grained recognition distinguishes among categories with subtle visual differences. In order to differentiate between these challenging visual categories, it is helpful to leverage additional information. Geolocation is a rich source of additional information that can be used to improve fine-grained classification accuracy, but has been understudied. Our contributions to this field are twofold. First, to the best of our knowledge, this is the first paper which systematically examined various ways of incorporating geolocation information into fine-grained image classification through the use of geolocation priors, post-processing or feature modulation. Secondly, to overcome the situation where no fine-grained dataset has complete geolocation information, we release two fine-grained datasets with geolocation by providing complementary information to existing popular datasets - iNaturalist and YFCC100M. By leveraging geolocation information we improve top-1 accuracy in iNaturalist from 70.1% to 79.0% for a strong baseline image-only model. Comparing several models, we found that best performance was achieved by a post-processing model that consumed the output of the image-only baseline alongside geolocation. However, for a resource-constrained model (MobileNetV2), performance was better with a feature modulation model that trains jointly over pixels and geolocation: accuracy increased from 59.6% to 72.2%. Our work makes a strong case for incorporating geolocation information in fine-grained recognition models for both server and on-device.

cs.CV↗

arXiv2019

A General Data Renewal Model for Prediction Algorithms in Industrial Data Analytics

In industrial data analytics, one of the fundamental problems is to utilize the temporal correlation of the industrial data to make timely predictions in the production process, such as fault prediction and yield prediction. However, the traditional prediction models are fixed while the conditions of the machines change over time, thus making the errors of predictions increase with the lapse of time. In this paper, we propose a general data renewal model to deal with it. Combined with the similarity function and the loss function, it estimates the time of updating the existing prediction model, then updates it according to the evaluation function iteratively and adaptively. We have applied the data renewal model to two prediction algorithms. The experiments demonstrate that the data renewal model can effectively identify the changes of data, update and optimize the prediction model so as to improve the accuracy of prediction.

cs.LG↗

arXiv2019

Unconventional crystal field splitting in non-centrosymmetric BaTiO$_3$ thin films

Understanding the crystal field splitting and orbital polarization in non-centrosymmetric systems such as ferroelectric materials is fundamentally important. In this study, taking BaTiO$_3$ (BTO) as a representative material we investigate titanium crystal field splitting and orbital polarization in non-centrosymmetric TiO$_6$ octahedra with resonant X-ray linear dichroism at Ti $L_{2,3}$-edge. The high-quality BaTiO$_3$ thin films were deposited on DyScO$_3$ (110) single crystal substrates in a layer-by-layer way by pulsed laser deposition. The reflection high-energy electron diffraction (RHEED) and element specific X-ray absorption spectroscopy (XAS) were performed to characterize the structural and electronic properties of the films. In sharp contrast to conventional crystal field splitting and orbital configuration ($d_{xz}$/$d_{yz}$ $<$ $d_{xy}$ $<$ $d_{3z^2-r^2}$ $<$ $d_{x^2-y^2}$ or $d_{xy}$ $<$ $d_{xz}$/$d_{yz}$ $<$ $d_{x^2-y^2}$ $<$ $d_{3z^2-r^2}$) according to Jahn-Teller effect, it is revealed that $d_{xz}$, $d_{yz}$, and $d_{xy}$ orbitals are nearly degenerate, whereas $d_{3z^2-r^2}$ and $d_{x^2-y^2}$ orbitals are split with an energy gap $\sim$ 100 meV in the epitaxial BTO films. The unexpected degenerate states $d_{xz}$/$d_{yz}$/$d_{xy}$ are coupled to Ti-O displacements resulting from competition between polar and Jahn-Teller distortions in non-centrosymmetric TiO$_6$ octhedra of BTO films. Our results provide a route to manipulate orbital degree of freedom by switching electric polarization in ferroelectric materials.

cond-mat.str-el↗

arXiv2019

LSTM-based Flow Prediction

In this paper, a method of prediction on continuous time series variables from the production or flow -- an LSTM algorithm based on multivariate tuning -- is proposed. The algorithm improves the traditional LSTM algorithm and converts the time series data into supervised learning sequences regarding industrial data's features. The main innovation of this paper consists in introducing the concepts of periodic measurement and time window in the industrial prediction problem, especially considering industrial data with time series characteristics. Experiments using real-world datasets show that the prediction accuracy is improved, 54.05% higher than that of traditional LSTM algorithm.

cs.LG↗

arXiv2019

Talking Face Generation by Conditional Recurrent Adversarial Network

Given an arbitrary face image and an arbitrary speech clip, the proposed work attempts to generating the talking face video with accurate lip synchronization while maintaining smooth transition of both lip and facial movement over the entire video clip. Existing works either do not consider temporal dependency on face images across different video frames thus easily yielding noticeable/abrupt facial and lip movement or are only limited to the generation of talking face video for a specific person thus lacking generalization capacity. We propose a novel conditional video generation network where the audio input is treated as a condition for the recurrent adversarial network such that temporal dependency is incorporated to realize smooth transition for the lip and facial movement. In addition, we deploy a multi-task adversarial training scheme in the context of video generation to improve both photo-realism and the accuracy for lip synchronization. Finally, based on the phoneme distribution information extracted from the audio clip, we develop a sample selection method that effectively reduces the size of the training dataset without sacrificing the quality of the generated video. Extensive experiments on both controlled and uncontrolled datasets demonstrate the superiority of the proposed approach in terms of visual quality, lip sync accuracy, and smooth transition of lip and facial movement, as compared to the state-of-the-art.

cs.CV↗

arXiv2019

Compressive Privacy for a Linear Dynamical System

We consider a linear dynamical system in which the state vector consists of both public and private states. One or more sensors make measurements of the state vector and sends information to a fusion center, which performs the final state estimation. To achieve an optimal tradeoff between the utility of estimating the public states and protection of the private states, the measurements at each time step are linearly compressed into a lower dimensional space. Under the centralized setting where all measurements are collected by a single sensor, we propose an optimization problem and an algorithm to find the best compression matrix. Under the decentralized setting where measurements are made separately at multiple sensors, each sensor optimizes its own local compression matrix. We propose methods to separate the overall optimization problem into multiple sub-problems that can be solved locally at each sensor. We consider the cases where there is no message exchange between the sensors; and where each sensor takes turns to transmit messages to the other sensors. Simulations and empirical experiments demonstrate the efficiency of our proposed approach in allowing the fusion center to estimate the public states with good accuracy while preventing it from estimating the private states accurately.

cs.IT↗

arXiv2019

Sliced Score Matching: A Scalable Approach to Density and Score Estimation

Score matching is a popular method for estimating unnormalized statistical models. However, it has been so far limited to simple, shallow models or low-dimensional data, due to the difficulty of computing the Hessian of log-density functions. We show this difficulty can be mitigated by projecting the scores onto random vectors before comparing them. This objective, called sliced score matching, only involves Hessian-vector products, which can be easily implemented using reverse-mode automatic differentiation. Therefore, sliced score matching is amenable to more complex models and higher dimensional data compared to score matching. Theoretically, we prove the consistency and asymptotic normality of sliced score matching estimators. Moreover, we demonstrate that sliced score matching can be used to learn deep score estimators for implicit distributions. In our experiments, we show sliced score matching can learn deep energy-based models effectively, and can produce accurate score estimates for applications such as variational inference with implicit distributions and training Wasserstein Auto-Encoders.

cs.LG↗

arXiv2019

Person Re-identification Using Visual Attention

Despite recent attempts for solving the person re-identification problem, it remains a challenging task since a person's appearance can vary significantly when large variations in view angle, human pose, and illumination are involved. In this paper, we propose a novel approach based on using a gradient-based attention mechanism in deep convolution neural network for solving the person re-identification problem. Our model learns to focus selectively on parts of the input image for which the networks' output is most sensitive to and processes them with high resolution while perceiving the surrounding image in low resolution. Extensive comparative evaluations demonstrate that the proposed method outperforms state-of-the-art approaches on the challenging CUHK01, CUHK03, and Market 1501 datasets.

cs.CV↗