arXiv subjects

Jing Ma

Publications and source records attributed to Jing Ma.

At least 163 records · Page 9Linked to original sources

arXiv2022

DecBERT: Enhancing the Language Understanding of BERT with Causal Attention Masks

Since 2017, the Transformer-based models play critical roles in various downstream Natural Language Processing tasks. However, a common limitation of the attention mechanism utilized in Transformer Encoder is that it cannot automatically capture the information of word order, so explicit position embeddings are generally required to be fed into the target model. In contrast, Transformer Decoder with the causal attention masks is naturally sensitive to the word order. In this work, we focus on improving the position encoding ability of BERT with the causal attention masks. Furthermore, we propose a new pre-trained language model DecBERT and evaluate it on the GLUE benchmark. Experimental results show that (1) the causal attention mask is effective for BERT on the language understanding tasks; (2) our DecBERT model without position embeddings achieve comparable performance on the GLUE benchmark; and (3) our modification accelerates the pre-training process and DecBERT w/ PE achieves better overall performance than the baseline systems when pre-training with the same amount of computational resources.

cs.CL↗

arXiv2022

A Frustratingly Simple Approach for End-to-End Image Captioning

Image Captioning is a fundamental task to join vision and language, concerning about cross-modal understanding and text generation. Recent years witness the emerging attention on image captioning. Most of existing works follow a traditional two-stage training paradigm. Before training the captioning models, an extra object detector is utilized to recognize the objects in the image at first. However, they require sizeable datasets with fine-grained object annotation for training the object detector, which is a daunting task. In addition, the errors of the object detectors are easy to propagate to the following captioning models, degenerating models' performance. To alleviate such defects, we propose a frustratingly simple but highly effective end-to-end image captioning framework, Visual Conditioned GPT (VC-GPT), by connecting the pre-trained visual encoder (CLIP-ViT) and language decoder (GPT2). Different from the vanilla connection method that directly inserts the cross-attention modules into GPT2, we come up with a self-ensemble cross-modal fusion mechanism that comprehensively considers both the single- and cross-modal knowledge. As a result, we do not need extra object detectors for model training. Experimental results conducted on three popular image captioning benchmarks (MSCOCO, Flickr30k and NoCaps) demonstrate that our VC-GPT achieves either the best or the second-best performance across all evaluation metrics over extensive baseline systems.

cs.CV↗

arXiv2022

Learning Fair Node Representations with Graph Counterfactual Fairness

Fair machine learning aims to mitigate the biases of model predictions against certain subpopulations regarding sensitive attributes such as race and gender. Among the many existing fairness notions, counterfactual fairness measures the model fairness from a causal perspective by comparing the predictions of each individual from the original data and the counterfactuals. In counterfactuals, the sensitive attribute values of this individual had been modified. Recently, a few works extend counterfactual fairness to graph data, but most of them neglect the following facts that can lead to biases: 1) the sensitive attributes of each node's neighbors may causally affect the prediction w.r.t. this node; 2) the sensitive attributes may causally affect other features and the graph structure. To tackle these issues, in this paper, we propose a novel fairness notion - graph counterfactual fairness, which considers the biases led by the above facts. To learn node representations towards graph counterfactual fairness, we propose a novel framework based on counterfactual data augmentation. In this framework, we generate counterfactuals corresponding to perturbations on each node's and their neighbors' sensitive attributes. Then we enforce fairness by minimizing the discrepancy between the representations learned from the original graph and the counterfactuals for each node. Experiments on both synthetic and real-world graphs show that our framework outperforms the state-of-the-art baselines in graph counterfactual fairness, and also achieves comparable prediction performance.

cs.LG↗

arXiv2021

Silting complexes and Gorenstein projective modules

We introduce Gorenstein silting modules (resp. complexes), and give the relation with the usual silting modules (resp. complexes). We show that Gorenstein silting modules are the module-theoretic counterpart of 2-term Gorenstein silting complexes; and partial Gorenstein silting modules are in bijection with τ_{G}-rigid modules for finite dimensional algebras of finite CM-type. We also give the relation between 2-term Gorenstein silting complexes, t-structures and torsion pair in module categories; and generalise the classical Brenner-Butler theorem to this setting; and characterise the global dimension of endomorphism algebras of 2-term Gorenstein silting complexes over an algebra A by terms of the Gorenstein global dimension of A.

math.AC↗

arXiv2021

Wavelength Conversion Efficiency Enhancement in Modal Phase Matched $χ^{(2)}$ Nonlinear Waveguides

Modal phase matching (MPM) is a widely used phase matching technique in Al$_x$Ga$_{1-x}$As and other $χ^{(2)}$ nonlinear waveguides for efficient wavelength conversions. The use of a non-fundamental spatial mode compensates the material dispersion but also reduces the spatial overlap of the three interacting waves and therefore limits the conversion efficiency. In this work, we develop a technique to increase the nonlinear overlap by modifying the material nonlinearity, instead of the traditional method of optimizing the modal field profiles. This could eliminate the limiting factor of low spatial overlap inherent to MPM and significantly enhance the conversion efficiency. Among the design examples provided, this technique could increase the conversion efficiency by a factor of up to $\sim$290 in an Al$_x$Ga$_{1-x}$As waveguide. We further show that this technique is applicable to all $χ^{(2)}$ material systems that utilize MPM for wavelength conversion.

physics.optics↗

arXiv2021

Spectrally pure photon pair generation in asymmetric heterogeneously coupled waveguides

In this work, we develop a design methodology to generate spectrally pure photon pairs in asymmetric heterogeneously coupled waveguides by spontaneous parametric down-conversion. Mode coupling in a system of waveguides is used to directly tailor the group velocity of a supermode to achieve group velocity matching (GVM) that is otherwise not allowed by material dispersion. Design examples based on thin film lithium niobate waveguides are provided, demonstrating high spectral purity and temperature tunability. This approach is a versatile strategy applicable to waveguides of different materials and structures, allowing more versatility in single-photon source designs.

physics.optics↗

arXiv2021

Coarse-To-Fine Incremental Few-Shot Learning

Different from fine-tuning models pre-trained on a large-scale dataset of preset classes, class-incremental learning (CIL) aims to recognize novel classes over time without forgetting pre-trained classes. However, a given model will be challenged by test images with finer-grained classes, e.g., a basenji is at most recognized as a dog. Such images form a new training set (i.e., support set) so that the incremental model is hoped to recognize a basenji (i.e., query) as a basenji next time. This paper formulates such a hybrid natural problem of coarse-to-fine few-shot (C2FS) recognition as a CIL problem named C2FSCIL, and proposes a simple, effective, and theoretically-sound strategy Knowe: to learn, normalize, and freeze a classifier's weights from fine labels, once learning an embedding space contrastively from coarse labels. Besides, as CIL aims at a stability-plasticity balance, new overall performance metrics are proposed. In that sense, on CIFAR-100, BREEDS, and tieredImageNet, Knowe outperforms all recent relevant CIL/FSCIL methods that are tailored to the new problem setting for the first time.

cs.CV↗

arXiv2021

Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks

Rumors are rampant in the era of social media. Conversation structures provide valuable clues to differentiate between real and fake claims. However, existing rumor detection methods are either limited to the strict relation of user responses or oversimplify the conversation structure. In this study, to substantially reinforces the interaction of user opinions while alleviating the negative impact imposed by irrelevant posts, we first represent the conversation thread as an undirected interaction graph. We then present a Claim-guided Hierarchical Graph Attention Network for rumor classification, which enhances the representation learning for responsive posts considering the entire social contexts and attends over the posts that can semantically infer the target claim. Extensive experiments on three Twitter datasets demonstrate that our rumor detection method achieves much better performance than state-of-the-art methods and exhibits a superior capacity for detecting rumors at early stages.

cs.CL↗

arXiv2021

Federated Graph Classification over Non-IID Graphs

Federated learning has emerged as an important paradigm for training machine learning models in different domains. For graph-level tasks such as graph classification, graphs can also be regarded as a special type of data samples, which can be collected and stored in separate local systems. Similar to other domains, multiple local systems, each holding a small set of graphs, may benefit from collaboratively training a powerful graph mining model, such as the popular graph neural networks (GNNs). To provide more motivation towards such endeavors, we analyze real-world graphs from different domains to confirm that they indeed share certain graph properties that are statistically significant compared with random graphs. However, we also find that different sets of graphs, even from the same domain or same dataset, are non-IID regarding both graph structures and node features. To handle this, we propose a graph clustered federated learning (GCFL) framework that dynamically finds clusters of local systems based on the gradients of GNNs, and theoretically justify that such clusters can reduce the structure and feature heterogeneity among graphs owned by the local systems. Moreover, we observe the gradients of GNNs to be rather fluctuating in GCFL which impedes high-quality clustering, and design a gradient sequence-based clustering mechanism based on dynamic time warping (GCFL+). Extensive experimental results and in-depth analysis demonstrate the effectiveness of our proposed frameworks.

cs.LG↗

arXiv2021

Malicious Mode Attack on EV Coordinated Charging Load and MIADRC Defense Strategy

The Internet of Things (IoT) provides a salient communication environment to facilitate the coordinated charging of electric vehicle (EV) load. However, as IoT is connected with the public network, the coordinated charging system is in a low-level cyber security and greatly vulnerable to malicious attacks. This paper investigates the malicious mode attack (MMA), which is a new cyberattack pattern that simultaneously attacks massive EV charging piles to generate continuous sinusoidal power disturbance with the same frequency as the poorly-damped wide-area electromechanical mode. Thereby, high amplitude forced oscillations could be stimulated by MMA, which seriously threats the power system stability. First, the potential threat of MMA is clarified by investigating the vulnerability of the IoT-based coordinated charging load control system, and an MMA process like Mirai is pointed out as an example. And then, based on the attack process, an MMA model is established for impact analysis where expressions of the mean and stochastic responses of the MMA forced oscillation are derived to discover main impact factors. Further, to mitigate the impact of MMA, a defense strategy based on multi-index information active disturbance rejection control is proposed to improve the stability and anti-disturbance ability of the power system, which considers the impact factors of both mode damping and disturbance compensation. Simulations are conducted to verify the existence and characteristics of MMA threats, and the efficiency of the proposed defense strategy is also validated.

eess.SY↗

arXiv2021

Cross-lingual COVID-19 Fake News Detection

The COVID-19 pandemic poses a great threat to global public health. Meanwhile, there is massive misinformation associated with the pandemic which advocates unfounded or unscientific claims. Even major social media and news outlets have made an extra effort in debunking COVID-19 misinformation, most of the fact-checking information is in English, whereas some unmoderated COVID-19 misinformation is still circulating in other languages, threatening the health of less-informed people in immigrant communities and developing countries. In this paper, we make the first attempt to detect COVID-19 misinformation in a low-resource language (Chinese) only using the fact-checked news in a high-resource language (English). We start by curating a Chinese real&fake news dataset according to existing fact-checking information. Then, we propose a deep learning framework named CrossFake to jointly encode the cross-lingual news body texts and capture the news content as much as possible. Empirical results on our dataset demonstrate the effectiveness of CrossFake under the cross-lingual setting and it also outperforms several monolingual and cross-lingual fake news detectors. The dataset is available at https://github.com/YingtongDou/CrossFake.

cs.CL↗

arXiv2021

Real-time Skeletonization for Sketch-based Modeling

Skeleton creation is an important phase in the character animation pipeline. However, handcrafting skeleton takes extensive labor time and domain knowledge. Automatic skeletonization provides a solution. However, most of the current approaches are far from real-time and lack the flexibility to control the skeleton complexity. In this paper, we present an efficient skeletonization method, which can be seamlessly integrated into the sketch-based modeling process in real-time. The method contains three steps: local sub-skeleton extraction; sub-skeleton connection; and global skeleton refinement. Firstly, the local skeleton is extracted from the processed polygon stroke and forms a subpart along with the sub-mesh. Then, local sub-skeletons are connected according to the intersecting relationships and the modeling sequence of subparts. Lastly, a global refinement method is proposed to give users coarse-to-fine control on the connected skeleton. We demonstrate the effectiveness of our method on a variety of examples created by both novices and professionals.

cs.GR↗

arXiv2021

Tailoring magnetic order via atomically stacking 3d/5d electrons

The ability to tune magnetic orders, such as magnetic anisotropy and topological spin texture, is desired in order to achieve high-performance spintronic devices. A recent strategy has been to employ interfacial engineering techniques, such as the introduction of spin-correlated interfacial coupling, to tailor magnetic orders and achieve novel magnetic properties. We chose a unique polar-nonpolar LaMnO3/SrIrO3 superlattice because Mn (3d)/Ir (5d) oxides exhibit rich magnetic behaviors and strong spin-orbit coupling through the entanglement of their 3d and 5d electrons. Through magnetization and magnetotransport measurements, we found that the magnetic order is interface-dominated as the superlattice period is decreased. We were able to then effectively modify the magnetization, tilt of the ferromagnetic easy axis, and symmetry transition of the anisotropic magnetoresistance of the LaMnO3/SrIrO3 superlattice by introducing additional Mn (3d) and Ir (5d) interfaces. Further investigations using in-depth first-principles calculations and numerical simulations revealed that these magnetic behaviors could be understood by the 3d/5d electron correlation and Rashba spin-orbit coupling. The results reported here demonstrate a new route to synchronously engineer magnetic properties through the atomic stacking of different electrons, contributing to future applications.

cond-mat.str-el↗

arXiv2021

On a sum involving certain arithmetic functions and the integral part function

In this short, we study sums of the shape $\sum_{n\leqslant x}{f([x/n])}/{[x/n]},$ where $f$ is Euler totient function $φ$, Dedekind function $Ψ$, sum-of-divisors function $σ$ or the alternating sum-of-divisors function $β.$ We improve previous results when $f=φ$ and derive new estimates when $f=Ψ, f=σ$ and $f=β.$

math.NT↗

arXiv2021

Temporal Network Embedding via Tensor Factorization

Representation learning on static graph-structured data has shown a significant impact on many real-world applications. However, less attention has been paid to the evolving nature of temporal networks, in which the edges are often changing over time. The embeddings of such temporal networks should encode both graph-structured information and the temporally evolving pattern. Existing approaches in learning temporally evolving network representations fail to capture the temporal interdependence. In this paper, we propose Toffee, a novel approach for temporal network representation learning based on tensor decomposition. Our method exploits the tensor-tensor product operator to encode the cross-time information, so that the periodic changes in the evolving networks can be captured. Experimental results demonstrate that Toffee outperforms existing methods on multiple real-world temporal networks in generating effective embeddings for the link prediction tasks.

cs.LG↗

arXiv2021

Multi-objective optimization and explanation for stroke risk assessment in Shanxi province

Stroke is the top leading causes of death in China (Zhou et al. The Lancet 2019). A dataset from Shanxi Province is used to identify the risk of each patient's at four states low/medium/high/attack and provide the state transition tendency through a SHAP DeepExplainer. To improve the accuracy on an imbalance sample set, the Quadratic Interactive Deep Neural Network (QIDNN) model is first proposed by flexible selecting and appending of quadratic interactive features. The experimental results showed that the QIDNN model with 7 interactive features achieve the state-of-art accuracy $83.25\%$. Blood pressure, physical inactivity, smoking, weight and total cholesterol are the top five important features. Then, for the sake of high recall on the most urgent state, attack state, the stroke occurrence prediction is taken as an auxiliary objective to benefit from multi-objective optimization. The prediction accuracy was promoted, meanwhile the recall of the attack state was improved by $24.9\%$ (to $84.83\%$) compared to QIDNN (from $67.93\%$) with same features. The prediction model and analysis tool in this paper not only gave the theoretical optimized prediction method, but also provided the attribution explanation of risk states and transition direction of each patient, which provided a favorable tool for doctors to analyze and diagnose the disease.

cs.LG↗

arXiv2021

Learning from Crowds by Modeling Common Confusions

Crowdsourcing provides a practical way to obtain large amounts of labeled data at a low cost. However, the annotation quality of annotators varies considerably, which imposes new challenges in learning a high-quality model from the crowdsourced annotations. In this work, we provide a new perspective to decompose annotation noise into common noise and individual noise and differentiate the source of confusion based on instance difficulty and annotator expertise on a per-instance-annotator basis. We realize this new crowdsourcing model by an end-to-end learning solution with two types of noise adaptation layers: one is shared across annotators to capture their commonly shared confusions, and the other one is pertaining to each annotator to realize individual confusion. To recognize the source of noise in each annotation, we use an auxiliary network to choose the two noise adaptation layers with respect to both instances and annotators. Extensive experiments on both synthesized and real-world benchmarks demonstrate the effectiveness of our proposed common noise adaptation solution.

cs.LG↗

arXiv2021

Analysis and classification of main risk factors causing stroke in Shanxi Province

In China, stroke is the first leading cause of death in recent years. It is a major cause of long-term physical and cognitive impairment, which bring great pressure on the National Public Health System. Evaluation of the risk of getting stroke is important for the prevention and treatment of stroke in China. A data set with 2000 hospitalized stroke patients in 2018 and 27583 residents during the year 2017 to 2020 is analyzed in this study. Due to data incompleteness, inconsistency, and non-structured formats, missing values in the raw data are filled with -1 as an abnormal class. With the cleaned features, three models on risk levels of getting stroke are built by using machine learning methods. The importance of "8+2" factors from China National Stroke Prevention Project (CSPP) is evaluated via decision tree and random forest models. Except for "8+2" factors the importance of features and SHAP1 values for lifestyle information, demographic information, and medical measurement are evaluated and ranked via a random forest model. Furthermore, a logistic regression model is applied to evaluate the probability of getting stroke for different risk levels. Based on the census data in both communities and hospitals from Shanxi Province, we investigate different risk factors of getting stroke and their ranking with interpretable machine learning models. The results show that Hypertension (Systolic blood pressure, Diastolic blood pressure), Physical Inactivity (Lack of sports), and Overweight (BMI) are ranked as the top three high-risk factors of getting stroke in Shanxi province. The probability of getting stroke for a person can also be predicted via our machine learning model.

cs.LG↗