Searcharxiv⌕ Search

arXiv subjects

Dongdong Zhang

Publications and source records attributed to Dongdong Zhang.

At least 91 records · Page 5Linked to original sources

Frobenius functors and Gorenstein projective precovers

We establish relations between Gorenstein projective precovers linked by Frobenius functors. This is motivated by an open problem that how to find general classes of rings for which modules have Gorenstein projective precovers. It is shown that if $F:\C\rightarrow\D$ is a separable Frobenius functor between abelian categories with enough projective objects, then every object in $\C$ has a Gorenstein projective precover provided that every object in $\D$ has a Gorenstein projective precover. This result is applied to separable Frobenius extensions and excellent extensions.

math.RA↗

Two new classes of n-exangulated categories

Herschend-Liu-Nakaoka introduced the notion of $n$-exangulated categories. It is not only a higher dimensional analogue of extriangulated categories defined by Nakaoka-Palu, but also gives a simultaneous generalization of $n$-exact categories and $(n+2)$-angulated categories. Let $\mathcal C$ be an $n$-exangulated category and $\mathcal X$ a full subcategory of $\mathcal C$. If $\mathcal X$ satisfies $\mathcal X\subseteq\mathcal P\cap\mathcal I$, then we give a necessary and sufficient condition for the ideal quotient $\mathcal C/\mathcal X$ to be an $n$-exangulated category, where $\mathcal P$ (resp. $\mathcal I$) is the full subcategory of projective (resp. injective) objects in $\mathcal C$. In addition, we define the notion of $n$-proper class in $\mathcal C$. If $ξ$ is an $n$-proper class in $\mathcal C$, then we prove that $\mathcal C$ admits a new $n$-exangulated structure. These two ways give $n$-exangulated categories which are neither $n$-exact nor $(n+2)$-angulated in general.

math.RT↗

Interpretable Deep Learning for Automatic Diagnosis of 12-lead Electrocardiogram

Electrocardiogram (ECG) is a widely used reliable, non-invasive approach for cardiovascular disease diagnosis. With the rapid growth of ECG examinations and the insufficiency of cardiologists, accurate and automatic diagnosis of ECG signals has become a hot research topic. Deep learning methods have demonstrated promising results in predictive healthcare tasks. In this paper, we developed a deep neural network for multi-label classification of cardiac arrhythmias in 12-lead ECG recordings. Experiments on a public 12-lead ECG dataset showed the effectiveness of our method. The proposed model achieved an average area under the receiver operating characteristic curve (AUC) of 0.970 and an average F1 score of 0.813. The deep model showed superior performance than 4 machine learning methods learned from extracted expert features. Besides, the deep models trained on single-lead ECGs produce lower performance than using all 12 leads simultaneously. The best-performing leads are lead I, aVR, and V5 among 12 leads. Finally, we employed the SHapley Additive exPlanations (SHAP) method to interpret the model's behavior at both patient level and population level. Our code is freely available at https://github.com/onlyzdd/ecg-diagnosis.

cs.LG↗

Document Classification for COVID-19 Literature

The global pandemic has made it more important than ever to quickly and accurately retrieve relevant scientific literature for effective consumption by researchers in a wide range of fields. We provide an analysis of several multi-label document classification models on the LitCovid dataset, a growing collection of 23,000 research papers regarding the novel 2019 coronavirus. We find that pre-trained language models fine-tuned on this dataset outperform all other baselines and that BioBERT surpasses the others by a small margin with micro-F1 and accuracy scores of around 86% and 75% respectively on the test set. We evaluate the data efficiency and generalizability of these models as essential features of any system prepared to deal with an urgent situation like the current health crisis. Finally, we explore 50 errors made by the best performing models on LitCovid documents and find that they often (1) correlate certain labels too closely together and (2) fail to focus on discriminative sections of the articles; both of which are important issues to address in future work. Both data and code are available on GitHub.

cs.IR↗

Avramov-Martsinkovsky type exact sequences for extriangulated categories

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a proper class $ξ$ of $\mathbb{E}$-triangles. In this paper, we first introduce the $ξ$-Gorenstein cohomology in terms of $ξ$-$\mathcal{G}$projective resolutions and $ξ$-$\mathcal{G}$injective coresolutions, respectively, and then we get the balance of $ξ$-Gorenstein cohomology. Moreover, we study the interplay among $ξ$-cohomology, $ξ$-Gorenstein cohomology and $ξ$-complete cohomology, and obtain the Avramov-Martsinkovsky type exact sequences in this setting.

math.CT↗

Syntax-aware Data Augmentation for Neural Machine Translation

Data augmentation is an effective performance enhancement in neural machine translation (NMT) by generating additional bilingual data. In this paper, we propose a novel data augmentation enhancement strategy for neural machine translation. Different from existing data augmentation methods which simply choose words with the same probability across different sentences for modification, we set sentence-specific probability for word selection by considering their roles in sentence. We use dependency parse tree of input sentence as an effective clue to determine selecting probability for every words in each sentence. Our proposed method is evaluated on WMT14 English-to-German dataset and IWSLT14 German-to-English dataset. The result of extensive experiments show our proposed syntax-aware data augmentation method may effectively boost existing sentence-independent methods for significant translation performance improvement.

cs.CL↗

Balance of complete cohomology in extriangulated categories

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a proper class $ξ$ of $\mathbb{E}$-triangles. In this paper, we study the balance of complete cohomology in $(\mathcal{C},\mathbb{E},\mathfrak{s})$, which is motivated by a result of Nucinkis that complete cohomology of modules is not balanced in the way the absolute cohomology Ext is balanced. As an application, we give some criteria for identifying a triangulated catgory to be Gorenstein and an artin algebra to be $F$-Gorenstein.

math.CT↗

Complete cohomology for extriangulated categories

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a proper class $ξ$ of $\mathbb{E}$-triangles. In this paper, we study complete cohomology of objects in $(\mathcal{C},\mathbb{E},\mathfrak{s})$ by applying $ξ$-projective resolutions and $ξ$-injective coresolutions constructed in $(\mathcal{C},\mathbb{E},\mathfrak{s})$. Vanishing of complete cohomology detects objects with finite $ξ$-projective dimension and finite $ξ$-injective dimension. As a consequence, we obtain some criteria for the validity of the Wakamatsu Tilting Conjecture and give a necessary and sufficient condition for a virtually Gorenstein algebra to be Gorenstein. Moreover, we give a general technique for computing complete cohomology of objects with finite $ξ$-$\mathcal{G}$projective dimension. As an application, the relationships between $ξ$-projective dimensions and $ξ$-$\mathcal{G}$projective dimensions for objects in $(\mathcal{C},\mathbb{E},\mathfrak{s})$ are given.

math.RT↗

XGPT: Cross-modal Generative Pre-Training for Image Captioning

While many BERT-based cross-modal pre-trained models produce excellent results on downstream understanding tasks like image-text retrieval and VQA, they cannot be applied to generation tasks directly. In this paper, we propose XGPT, a new method of Cross-modal Generative Pre-Training for Image Captioning that is designed to pre-train text-to-image caption generators through three novel generation tasks, including Image-conditioned Masked Language Modeling (IMLM), Image-conditioned Denoising Autoencoding (IDA), and Text-conditioned Image Feature Generation (TIFG). As a result, the pre-trained XGPT can be fine-tuned without any task-specific architecture modifications to create state-of-the-art models for image captioning. Experiments show that XGPT obtains new state-of-the-art results on the benchmark datasets, including COCO Captions and Flickr30k Captions. We also use XGPT to generate new image captions as data augmentation for the image retrieval task and achieve significant improvement on all recall metrics.

cs.CL↗

Proper resolutions and Gorensteinness in extriangulated categories

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a proper class $ξ$ of $\mathbb{E}$-triangles, and $\mathcal{W}$ an additive full subcategory of $\mathcal C$. We provide a method for constructing a proper $\mathcal{W}(ξ)$-resolution (respectively, coproper $\mathcal{W}(ξ)$-coresolution) of one term in an $\mathbb{E}$-triangle in $ξ$ from that of the other two terms. By using this way, we establish the stability of the Gorenstein category $\mathcal{GW}(ξ)$ in extriangulated categories. These results generalise their work by Huang and Yang-Wang, but the proof is not too far from their case. Finally, we give some applications about our main results.

math.RT↗

A proximal MM method for the zero-norm regularized PLQ composite optimization problem

This paper is concerned with a class of zero-norm regularized piecewise linear-quadratic (PLQ) composite minimization problems, which covers the zero-norm regularized $\ell_1$-loss minimization problem as a special case. For this class of nonconvex nonsmooth problems, we show that its equivalent MPEC reformulation is partially calm on the set of global optima and make use of this property to derive a family of equivalent DC surrogates. Then, we propose a proximal majorization-minimization (MM) method, a convex relaxation approach not in the DC algorithm framework, for solving one of the DC surrogates which is a semiconvex PLQ minimization problem involving three nonsmooth terms. For this method, we establish its global convergence and linear rate of convergence, and under suitable conditions show that the limit of the generated sequence is not only a local optimum but also a good critical point in a statistical sense. Numerical experiments are conducted with synthetic and real data for the proximal MM method with the subproblems solved by a dual semismooth Newton method to confirm our theoretical findings, and numerical comparisons with a convergent indefinite-proximal ADMM for the partially smoothed DC surrogate verify its superiority in the quality of solutions and computing time.

math.OC↗

Proper classes and Gorensteinness in extriangulated categories

Extriangulated categories were introduced by Nakaoka and Palu as a simultaneous generalization of exact categories and triangulated categories. A notion of proper class in an extriangulated category is defined in this paper. Let $\mathcal{C}$ be an extriangulated category and $ξ$ a proper class in $\mathcal{C}$. We prove that $\mathcal{C}$ admits a new extriangulated structure. This construction gives extriangulated categories which are neither exact categories nor triangulated categories. Moreover, we introduce and study $ξ$-Gorenstein projective objects in $\mathcal{C}$ and demonstrate that $ξ$-Gorenstein projective objects share some basic properties with Gorenstein projective objects in module categories or in triangulated categories. In particular, we refine a result of Asadollahi and Salarian [Gorenstein objects in triangulated categories, J. Algebra 281(2004), 264-286]. As an application, the $ξ$-$\mathcal{G}$projective model structures on extriangulated categories are obtained.

math.RT↗

Deep Learning Decoding of Mental State in Non-invasive Brain Computer Interface

Brain computer interface (BCI) has been popular as a key approach to monitor our brains recent year. Mental states monitoring is one of the most important BCI applications and becomes increasingly accessible. However, the mental state prediction accuracy and generality through encephalogram (EEG) are not good enough for everyday use. Here in this paper we present a deep learning-based EEG decoding method to read mental states. We propose a novel 1D convolutional neural network with different filter lengths to capture different frequency bands information. To improve the prediction accuracy, we also used a resnet-like structure to train a relatively deep convolutional neural network to promote feature extraction. Compared with traditional ways of predictions such as KNN and SVM, we achieved a significantly better result with an accuracy of 96.40%. Also, in contrast with some already published open source deep neural network structures, our methods achieved the state of art prediction accuracy on a mental state recognition dataset. Our results demonstrate using only 1D convolution could extract the features of EEG and the possibility of mental state prediction using portable EEG devices.

eess.SP↗

Cross-Enhancement Transform Two-Stream 3D ConvNets for Action Recognition

Action recognition is an important research topic in computer vision. It is the basic work for visual understanding and has been applied in many fields. Since human actions can vary in different environments, it is difficult to infer actions in completely different states with a same structural model. For this case, we propose a Cross-Enhancement Transform Two-Stream 3D ConvNets algorithm, which considers the action distribution characteristics on the specific dataset. As a teaching model, stream with better performance in both streams is expected to assist in training another stream. In this way, the enhanced-trained stream and teacher stream are combined to infer actions. We implement experiments on the video datasets UCF-101, HMDB-51, and Kinetics-400, and the results confirm the effectiveness of our algorithm.

cs.CV↗

Long-term trapping of cold polar molecules

We demonstrate the long-term ($<$ 1 minute) trapping of Stark-decelerated OH radicals in their $X~^{2}Π_{3/2}~(ν= 0,~J = 3/2,~M_{J} = 3/2,~f)$ state in a permanent magnetic trap. The trap environment was cryogenically cooled to a temperature of 17 K in order to efficiently suppress black-body-radiation-induced pumping of the molecules out of trappable quantum states and collisions with residual background gas molecules which usually limit the trap lifetimes. The cold molecules were kept confined on timescales approaching minutes, an improvement of up to two orders of magnitude compared to room-temperature experiments, at translational temperatures on the order of 25 mK. The present results pave the way for spectroscopic studies of trapped molecules with long interaction times enabling high precision, for investigations of cold collisions and reactions with very small reaction rates, for new avenues for the production of ultracold molecules via sympathetic cooling and for the realisation of new forms of hybrid matter with co-trapped atoms or ions.

physics.chem-ph↗

Gorenstein homological dimensions for extriangulated categories

Let $(\mathcal{C},\mathbb{E},\mathfrak{s})$ be an extriangulated category with a proper class $ξ$ of $\mathbb{E}$-triangles. The authors introduced and studied $ξ$-$\mathcal{G}$projective and $ξ$-$\mathcal{G}$injective in \cite{HZZ}. In this paper, we discuss Gorenstein homological dimensions for extriangulated categories. More precisely, we first give some characterizations of $ξ$-$\mathcal{G}$projective dimension by using derived functors on $\mathcal{C}$. Second, let $\mathcal{P}(ξ)$ (resp. $\mathcal{I}(ξ)$) be a generating (resp. cogenerating) subcategory of $\mathcal{C}$. We show that the following equality holds under some assumptions: $$\sup\{ξ\textrm{-}\mathcal{G}{\rm pd}M \ | \ \textrm{for} \ \textrm{any} \ M\in{\mathcal{C}}\}=\sup\{ξ\textrm{-}\mathcal{G}{\rm id}M \ | \ \textrm{for} \ \textrm{any} \ M\in{\mathcal{C}}\},$$ where $ξ\textrm{-}\mathcal{G}{\rm pd}M$ (resp. $ξ\textrm{-}\mathcal{G}{\rm id}M$) denotes $ξ$-$\mathcal{G}$projective (resp. $ξ$-$\mathcal{G}$injective) dimension of $M$. As an application, our main results generalize their work by Bennis-Mahdou and Ren-Liu. Moreover, our proof is not far from the usual module or triangulated case.

math.RT↗

A Novel Task-Oriented Text Corpus in Silent Speech Recognition and its Natural Language Generation Construction Method

Millions of people with severe speech disorders around the world may regain their communication capabilities through techniques of silent speech recognition (SSR). Using electroencephalography (EEG) as a biomarker for speech decoding has been popular for SSR. However, the lack of SSR text corpus has impeded the development of this technique. Here, we construct a novel task-oriented text corpus, which is utilized in the field of SSR. In the process of construction, we propose a task-oriented hybrid construction method based on natural language generation algorithm. The algorithm focuses on the strategy of data-to-text generation, and has two advantages including linguistic quality and high diversity. These two advantages use template-based method and deep neural networks respectively. In an SSR experiment with the generated text corpus, analysis results show that the performance of our hybrid construction method outperforms the pure method such as template-based natural language generation or neural natural language generation models.

cs.CL↗

Learning Unsupervised Word Mapping by Maximizing Mean Discrepancy

Cross-lingual word embeddings aim to capture common linguistic regularities of different languages, which benefit various downstream tasks ranging from machine translation to transfer learning. Recently, it has been shown that these embeddings can be effectively learned by aligning two disjoint monolingual vector spaces through a linear transformation (word mapping). In this work, we focus on learning such a word mapping without any supervision signal. Most previous work of this task adopts parametric metrics to measure distribution differences, which typically requires a sophisticated alternate optimization process, either in the form of \emph{minmax game} or intermediate \emph{density estimation}. This alternate optimization process is relatively hard and unstable. In order to avoid such sophisticated alternate optimization, we propose to learn unsupervised word mapping by directly maximizing the mean discrepancy between the distribution of transferred embedding and target embedding. Extensive experimental results show that our proposed model outperforms competitive baselines by a large margin.

cs.CL↗