arXiv subjects

Cheng Yang

Publications and source records attributed to Cheng Yang.

At least 217 records · Page 12Linked to original sources

arXiv2021

From Symbols to Embeddings: A Tale of Two Representations in Computational Social Science

Computational Social Science (CSS), aiming at utilizing computational methods to address social science problems, is a recent emerging and fast-developing field. The study of CSS is data-driven and significantly benefits from the availability of online user-generated contents and social networks, which contain rich text and network data for investigation. However, these large-scale and multi-modal data also present researchers with a great challenge: how to represent data effectively to mine the meanings we want in CSS? To explore the answer, we give a thorough review of data representations in CSS for both text and network. Specifically, we summarize existing representations into two schemes, namely symbol-based and embedding-based representations, and introduce a series of typical methods for each scheme. Afterwards, we present the applications of the above representations based on the investigation of more than 400 research articles from 6 top venues involved with CSS. From the statistics of these applications, we unearth the strength of each kind of representations and discover the tendency that embedding-based representations are emerging and obtaining increasing attention over the last decade. Finally, we discuss several key challenges and open issues for future directions. This survey aims to provide a deeper understanding and more advisable applications of data representations for CSS researchers.

cs.CY↗

arXiv2021

Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment

Given a convex and differentiable objective $Q(\M)$ for a real symmetric matrix $\M$ in the positive definite (PD) cone -- used to compute Mahalanobis distances -- we propose a fast general metric learning framework that is entirely projection-free. We first assume that $\M$ resides in a space $\cS$ of generalized graph Laplacian matrices corresponding to balanced signed graphs. $\M \in \cS$ that is also PD is called a graph metric matrix. Unlike low-rank metric matrices common in the literature, $\cS$ includes the important diagonal-only matrices as a special case. The key theorem to circumvent full eigen-decomposition and enable fast metric matrix optimization is Gershgorin disc perfect alignment (GDPA): given $\M \in \cS$ and diagonal matrix $§$, where $S_{ii} = 1/v_i$ and $\v$ is $\M$'s first eigenvector, we prove that Gershgorin disc left-ends of similarity transform $\B = §\M §^{-1}$ are perfectly aligned at the smallest eigenvalue $λ_{\min}$. Using this theorem, we replace the PD cone constraint in the metric learning problem with tightest possible linear constraints per iteration, so that the alternating optimization of the diagonal / off-diagonal terms in $\M$ can be solved efficiently as linear programs via the Frank-Wolfe method. We update $\v$ using Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) with warm start as entries in $\M$ are optimized successively. Experiments show that our graph metric optimization is significantly faster than cone-projection schemes, and produces competitive binary classification performance.

cs.LG↗

arXiv2021

KACC: A Multi-task Benchmark for Knowledge Abstraction, Concretization and Completion

A comprehensive knowledge graph (KG) contains an instance-level entity graph and an ontology-level concept graph. The two-view KG provides a testbed for models to "simulate" human's abilities on knowledge abstraction, concretization, and completion (KACC), which are crucial for human to recognize the world and manage learned knowledge. Existing studies mainly focus on partial aspects of KACC. In order to promote thorough analyses for KACC abilities of models, we propose a unified KG benchmark by improving existing benchmarks in terms of dataset scale, task coverage, and difficulty. Specifically, we collect new datasets that contain larger concept graphs, abundant cross-view links as well as dense entity graphs. Based on the datasets, we propose novel tasks such as multi-hop knowledge abstraction (MKA), multi-hop knowledge concretization (MKC) and then design a comprehensive benchmark. For MKA and MKC tasks, we further annotate multi-hop hierarchical triples as harder samples. The experimental results of existing methods demonstrate the challenges of our benchmark. The resource is available at https://github.com/thunlp/KACC.

cs.CL↗

arXiv2021

Markdowns in E-Commerce Fresh Retail: A Counterfactual Prediction and Multi-Period Optimization Approach

In this paper, by leveraging abundant observational transaction data, we propose a novel data-driven and interpretable pricing approach for markdowns, consisting of counterfactual prediction and multi-period price optimization. Firstly, we build a semi-parametric structural model to learn individual price elasticity and predict counterfactual demand. This semi-parametric model takes advantage of both the predictability of nonparametric machine learning model and the interpretability of economic model. Secondly, we propose a multi-period dynamic pricing algorithm to maximize the overall profit of a perishable product over its finite selling horizon. Different with the traditional approaches that use the deterministic demand, we model the uncertainty of counterfactual demand since it inevitably has randomness in the prediction process. Based on the stochastic model, we derive a sequential pricing strategy by Markov decision process, and design a two-stage algorithm to solve it. The proposed algorithm is very efficient. It reduces the time complexity from exponential to polynomial. Experimental results show the advantages of our pricing algorithm, and the proposed framework has been successfully deployed to the well-known e-commerce fresh retail scenario - Freshippo.

cs.AI↗

arXiv2021

Phonon-related monochromatic THz radiation and its magneto-modulation in 2D ferromagnetic Cr2Ge2Te6

Searching multiple types of terahertz (THz) irradiation source is crucial for the THz technology. Here, by utilizing a two-dimensional (2D) ferromagnetic Cr2Ge2Te6 crystal, we firstly demonstrate a magneto-tunable monochromatic THz irradiation source. With a low-photonic-energy broadband THz pump, a strong THz irradiation with frequency ~0.9 THz and bandwidth ~0.25 THz can be generated and its conversion efficiency could even reach 2.1% at 160 K. Moreover, it is intriguing to find that such monochromatic THz irradiation can be efficiently modulated by the magnetic field below 160 K. According to both experimental and theoretical analyses, the emergent THz irradiation is identified as the emission from the phonon-polariton and its temperature and magnetic field dependent behaviors confirmed the large spin-lattice coupling in this 2D ferromagnetic crystal. These observations provide a new route for the creation of tunable monochromatic THz source which may have great practical interests in future applications in photonic and spintronic devices.

physics.optics↗

arXiv2021

Vortex motion of the Euler and Lake equations

We start by surveying the planar point vortex motion of the Euler equations in the whole plane, half-plane and quadrant. Then we go on to prove non-collision property of 2-vortex system by using the explicit form of orbits of 2-vortex system in the half-plane. We also prove that the $N$-vortex system in the half-plane is nonintegrable for $N>2$, which is suggested previously by numerical experiments without rigorous proof. The skew-mean-curvature (or binormal) flow in $\mathbb R^n,\;n\geq3$ with certain symmetry can be regarded as point vortex motion of the 2D lake equations. We compare point vortex motions of the Euler and lake equations. Interesting similarities between the point vortex motion in the half-plane, quadrant and the binormal motion of coaxial vortex rings, sphere product membranes are addressed. We also raise some open questions in the paper.

math-ph↗

arXiv2021

DCNAS: Densely Connected Neural Architecture Search for Semantic Image Segmentation

Neural Architecture Search (NAS) has shown great potentials in automatically designing scalable network architectures for dense image predictions. However, existing NAS algorithms usually compromise on restricted search space and search on proxy task to meet the achievable computational demands. To allow as wide as possible network architectures and avoid the gap between target and proxy dataset, we propose a Densely Connected NAS (DCNAS) framework, which directly searches the optimal network structures for the multi-scale representations of visual information, over a large-scale target dataset. Specifically, by connecting cells with each other using learnable weights, we introduce a densely connected search space to cover an abundance of mainstream network designs. Moreover, by combining both path-level and channel-level sampling strategies, we design a fusion module to reduce the memory consumption of ample search space. We demonstrate that the architecture obtained from our DCNAS algorithm achieves state-of-the-art performances on public semantic image segmentation benchmarks, including 84.3% on Cityscapes, and 86.9% on PASCAL VOC 2012. We also retain leading performances when evaluating the architecture on the more challenging ADE20K and Pascal Context dataset.

cs.CV↗

arXiv2021

Extract the Knowledge of Graph Neural Networks and Go Beyond it: An Effective Knowledge Distillation Framework

Semi-supervised learning on graphs is an important problem in the machine learning area. In recent years, state-of-the-art classification methods based on graph neural networks (GNNs) have shown their superiority over traditional ones such as label propagation. However, the sophisticated architectures of these neural models will lead to a complex prediction mechanism, which could not make full use of valuable prior knowledge lying in the data, e.g., structurally correlated nodes tend to have the same class. In this paper, we propose a framework based on knowledge distillation to address the above issues. Our framework extracts the knowledge of an arbitrary learned GNN model (teacher model), and injects it into a well-designed student model. The student model is built with two simple prediction mechanisms, i.e., label propagation and feature transformation, which naturally preserves structure-based and feature-based prior knowledge, respectively. In specific, we design the student model as a trainable combination of parameterized label propagation and feature transformation modules. As a result, the learned student can benefit from both prior knowledge and the knowledge in GNN teachers for more effective predictions. Moreover, the learned student model has a more interpretable prediction process than GNNs. We conduct experiments on five public benchmark datasets and employ seven GNN models including GCN, GAT, APPNP, SAGE, SGC, GCNII and GLP as the teacher models. Experimental results show that the learned student model can consistently outperform its corresponding teacher model by 1.4% - 4.7% on average. Code and data are available at https://github.com/BUPT-GAMMA/CPF

cs.LG↗

arXiv2021

Re-rank Coarse Classification with Local Region Enhanced Features for Fine-Grained Image Recognition

Fine-grained image recognition is very challenging due to the difficulty of capturing both semantic global features and discriminative local features. Meanwhile, these two features are not easy to be integrated, which are even conflicting when used simultaneously. In this paper, a retrieval-based coarse-to-fine framework is proposed, where we re-rank the TopN classification results by using the local region enhanced embedding features to improve the Top1 accuracy (based on the observation that the correct category usually resides in TopN results). To obtain the discriminative regions for distinguishing the fine-grained images, we introduce a weakly-supervised method to train a box generating branch with only image-level labels. In addition, to learn more effective semantic global features, we design a multi-level loss over an automatically constructed hierarchical category structure. Experimental results show that our method achieves state-of-the-art performance on three benchmarks: CUB-200-2011, Stanford Cars, and FGVC Aircraft. Also, visualizations and analysis are provided for better understanding.

cs.CV↗

arXiv2021

CoRe: An Efficient Coarse-refined Training Framework for BERT

In recent years, BERT has made significant breakthroughs on many natural language processing tasks and attracted great attentions. Despite its accuracy gains, the BERT model generally involves a huge number of parameters and needs to be trained on massive datasets, so training such a model is computationally very challenging and time-consuming. Hence, training efficiency should be a critical issue. In this paper, we propose a novel coarse-refined training framework named CoRe to speed up the training of BERT. Specifically, we decompose the training process of BERT into two phases. In the first phase, by introducing fast attention mechanism and decomposing the large parameters in the feed-forward network sub-layer, we construct a relaxed BERT model which has much less parameters and much lower model complexity than the original BERT, so the relaxed model can be quickly trained. In the second phase, we transform the trained relaxed BERT model into the original BERT and further retrain the model. Thanks to the desired initialization provided by the relaxed model, the retraining phase requires much less training steps, compared with training an original BERT model from scratch with a random initialization. Experimental results show that the proposed CoRe framework can greatly reduce the training time without reducing the performance.

cs.CL↗

arXiv2021

Incorporating Vision Bias into Click Models for Image-oriented Search Engine

Most typical click models assume that the probability of a document to be examined by users only depends on position, such as PBM and UBM. It works well in various kinds of search engines. However, in a search engine where massive candidate documents display images as responses to the query, the examination probability should not only depend on position. The visual appearance of an image-oriented document also plays an important role in its opportunity to be examined. In this paper, we assume that vision bias exists in an image-oriented search engine as another crucial factor affecting the examination probability aside from position. Specifically, we apply this assumption to classical click models and propose an extended model, to better capture the examination probabilities of documents. We use regression-based EM algorithm to predict the vision bias given the visual features extracted from candidate documents. Empirically, we evaluate our model on a dataset developed from a real-world online image-oriented search engine, and demonstrate that our proposed model can achieve significant improvements over its baseline model in data fitness and sparsity handling.

cs.AI↗

arXiv2020

Two-dimensional Cold Electron Transport for Steep-slope Transistors

Room-temperature Fermi-Dirac electron thermal excitation in conventional three-dimensional (3D) or two-dimensional (2D) semiconductors generates hot electrons with a relatively long thermal tail in energy distribution. These hot electrons set a fundamental obstacle known as the "Boltzmann tyranny" that limits the subthreshold swing (SS) and therefore the minimum power consumption of 3D and 2D field-effect transistors (FETs). Here, we investigated a novel graphene (Gr)-enabled cold electron injection where the Gr acts as the Dirac source to provide the cold electrons with a localized electron density distribution and a short thermal tail at room temperature. These cold electrons correspond to an electronic cooling effect with the effective electron temperature of ~145 K in the monolayer MoS2, which enable the transport factor lowering and thus the steep-slope switching (across for 3 decades with a minimum SS of 29 mV/decade at room temperature) for a monolayer MoS2 FET. Especially, a record-high sub-60-mV/decade current density (over 1 μA/μm) can be achieved compared to conventional steep-slope technologies such as tunneling FETs or negative capacitance FETs using 2D or 3D channel materials. Our work demonstrates the great potential of 2D Dirac-source cold electron transistor as an innovative steep-slope transistor concept, and provides new opportunities for 2D materials toward future energy-efficient nanoelectronics.

cond-mat.mes-hall↗

arXiv2020

Information Diffusion Prediction with Latent Factor Disentanglement

Information diffusion prediction is a fundamental task which forecasts how an information item will spread among users. In recent years, deep learning based methods, especially those based on recurrent neural networks (RNNs), have achieved promising results on this task by treating infected users as sequential data. However, existing methods represent all previously infected users by a single vector and could fail to encode all necessary information for future predictions due to the mode collapse problem. To address this problem, we propose to employ the idea of disentangled representation learning, which aims to extract multiple latent factors representing different aspects of the data, for modeling the information diffusion process. Specifically, we employ a sequential attention module and a disentangled attention module to better aggregate the history information and disentangle the latent factors. Experimental results on three real-world datasets show that the proposed model SIDDA significantly outperforms state-of-the-art baseline methods by up to 14% in terms of hits@N metric, which demonstrates the effectiveness of our method.

cs.SI↗

arXiv2020

Progressively Stacking 2.0: A Multi-stage Layerwise Training Method for BERT Training Speedup

Pre-trained language models, such as BERT, have achieved significant accuracy gain in many natural language processing tasks. Despite its effectiveness, the huge number of parameters makes training a BERT model computationally very challenging. In this paper, we propose an efficient multi-stage layerwise training (MSLT) approach to reduce the training time of BERT. We decompose the whole training process into several stages. The training is started from a small model with only a few encoder layers and we gradually increase the depth of the model by adding new encoder layers. At each stage, we only train the top (near the output layer) few encoder layers which are newly added. The parameters of the other layers which have been trained in the previous stages will not be updated in the current stage. In BERT training, the backward computation is much more time-consuming than the forward computation, especially in the distributed training setting in which the backward computation time further includes the communication time for gradient synchronization. In the proposed training strategy, only top few layers participate in backward computation, while most layers only participate in forward computation. Hence both the computation and communication efficiencies are greatly improved. Experimental results show that the proposed method can achieve more than 110% training speedup without significant performance degradation.

cs.CL↗

arXiv2020

A Multi-Stage Adaptive Sampling Scheme for Passivity Characterization of Large-Scale Macromodels

This paper proposes a hierarchical adaptive sampling scheme for passivity characterization of large-scale linear lumped macromodels. Here, large-scale is intended both in terms of dynamic order and especially number of input/output ports. Standard passivity characterization approaches based on spectral properties of associated Hamiltonian matrices are either inefficient or non-applicable for large-scale models, due to an excessive computational cost. This paper builds on existing adaptive sampling methods and proposes a hybrid multi-stage algorithm that is able to detect the passivity violations with limited computing resources. Results from extensive testing demonstrate a major reduction in computational requirements with respect to competing approaches.

cs.CE↗

arXiv2020

Identification of deep breath while moving forward based on multiple body regions and graph signal analysis

This paper presents an unobtrusive solution that can automatically identify deep breath when a person is walking past the global depth camera. Existing non-contact breath assessments achieve satisfactory results under restricted conditions when human body stays relatively still. When someone moves forward, the breath signals detected by depth camera are hidden within signals of trunk displacement and deformation, and the signal length is short due to the short stay time, posing great challenges for us to establish models. To overcome these challenges, multiple region of interests (ROIs) based signal extraction and selection method is proposed to automatically obtain the signal informative to breath from depth video. Subsequently, graph signal analysis (GSA) is adopted as a spatial-temporal filter to wipe the components unrelated to breath. Finally, a classifier for identifying deep breath is established based on the selected breath-informative signal. In validation experiments, the proposed approach outperforms the comparative methods with the accuracy, precision, recall and F1 of 75.5%, 76.2%, 75.0% and 75.2%, respectively. This system can be extended to public places to provide timely and ubiquitous help for those who may have or are going through physical or mental trouble.

cs.CV↗

arXiv2020

AP20-OLR Challenge: Three Tasks and Their Baselines

This paper introduces the fifth oriental language recognition (OLR) challenge AP20-OLR, which intends to improve the performance of language recognition systems, along with APSIPA Annual Summit and Conference (APSIPA ASC). The data profile, three tasks, the corresponding baselines, and the evaluation principles are introduced in this paper. The AP20-OLR challenge includes more languages, dialects and real-life data provided by Speechocean and the NSFC M2ASR project, and all the data is free for participants. The challenge this year still focuses on practical and challenging problems, with three tasks: (1) cross-channel LID, (2) dialect identification and (3) noisy LID. Based on Kaldi and Pytorch, recipes for i-vector and x-vector systems are also conducted as baselines for the three tasks. These recipes will be online-published, and available for participants to configure LID systems. The baseline results on the three tasks demonstrate that those tasks in this challenge are worth paying more efforts to achieve better performance.

eess.AS↗

arXiv2020

Improving Sequence Modeling Ability of Recurrent Neural Networks via Sememes

Sememes, the minimum semantic units of human languages, have been successfully utilized in various natural language processing applications. However, most existing studies exploit sememes in specific tasks and few efforts are made to utilize sememes more fundamentally. In this paper, we propose to incorporate sememes into recurrent neural networks (RNNs) to improve their sequence modeling ability, which is beneficial to all kinds of downstream tasks. We design three different sememe incorporation methods and employ them in typical RNNs including LSTM, GRU and their bidirectional variants. In evaluation, we use several benchmark datasets involving PTB and WikiText-2 for language modeling, SNLI for natural language inference and another two datasets for sentiment analysis and paraphrase detection. Experimental results show evident and consistent improvement of our sememe-incorporated models compared with vanilla RNNs, which proves the effectiveness of our sememe incorporation methods. Moreover, we find the sememe-incorporated models have higher robustness and outperform adversarial training in defending adversarial attack. All the code and data of this work can be obtained at https://github.com/thunlp/SememeRNN.

cs.CL↗