arXiv subjects

Wei Ye

Publications and source records attributed to Wei Ye.

At least 181 records · Page 10Linked to original sources

arXiv2022

Learning Deep Graph Representations via Convolutional Neural Networks

Graph-structured data arise in many scenarios. A fundamental problem is to quantify the similarities of graphs for tasks such as classification. R-convolution graph kernels are positive-semidefinite functions that decompose graphs into substructures and compare them. One problem in the effective implementation of this idea is that the substructures are not independent, which leads to high-dimensional feature space. In addition, graph kernels cannot capture the high-order complex interactions between vertices. To mitigate these two problems, we propose a framework called DeepMap to learn deep representations for graph feature maps. The learned deep representation for a graph is a dense and low-dimensional vector that captures complex high-order interactions in a vertex neighborhood. DeepMap extends Convolutional Neural Networks (CNNs) to arbitrary graphs by generating aligned vertex sequences and building the receptive field for each vertex. We empirically validate DeepMap on various graph classification benchmarks and demonstrate that it achieves state-of-the-art performance.

cs.LG↗

arXiv2022

Modeling Human-AI Team Decision Making

AI and humans bring complementary skills to group deliberations. Modeling this group decision making is especially challenging when the deliberations include an element of risk and an exploration-exploitation process of appraising the capabilities of the human and AI agents. To investigate this question, we presented a sequence of intellective issues to a set of human groups aided by imperfect AI agents. A group's goal was to appraise the relative expertise of the group's members and its available AI agents, evaluate the risks associated with different actions, and maximize the overall reward by reaching consensus. We propose and empirically validate models of human-AI team decision making under such uncertain circumstances, and show the value of socio-cognitive constructs of prospect theory, influence dynamics, and Bayesian learning in predicting the behavior of human-AI groups.

cs.HC↗

arXiv2021

Frequency-Aware Contrastive Learning for Neural Machine Translation

Low-frequency word prediction remains a challenge in modern neural machine translation (NMT) systems. Recent adaptive training methods promote the output of infrequent words by emphasizing their weights in the overall training objectives. Despite the improved recall of low-frequency words, their prediction precision is unexpectedly hindered by the adaptive objectives. Inspired by the observation that low-frequency words form a more compact embedding space, we tackle this challenge from a representation learning perspective. Specifically, we propose a frequency-aware token-level contrastive learning method, in which the hidden state of each decoding step is pushed away from the counterparts of other target words, in a soft contrastive way based on the corresponding word frequencies. We conduct experiments on widely used NIST Chinese-English and WMT14 English-German translation tasks. Empirical results show that our proposed methods can not only significantly improve the translation quality but also enhance lexical diversity and optimize word representation space. Further investigation reveals that, comparing with related adaptive training strategies, the superiority of our method on low-frequency word prediction lies in the robustness of token-level recall across different frequencies without sacrificing precision.

cs.CL↗

arXiv2021

Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation

Vision Transformers (ViTs) have emerged with superior performance on computer vision tasks compared to convolutional neural network (CNN)-based models. However, ViTs are mainly designed for image classification that generate single-scale low-resolution representations, which makes dense prediction tasks such as semantic segmentation challenging for ViTs. Therefore, we propose HRViT, which enhances ViTs to learn semantically-rich and spatially-precise multi-scale representations by integrating high-resolution multi-branch architectures with ViTs. We balance the model performance and efficiency of HRViT by various branch-block co-optimization techniques. Specifically, we explore heterogeneous branch designs, reduce the redundancy in linear layers, and augment the attention block with enhanced expressiveness. Those approaches enabled HRViT to push the Pareto frontier of performance and efficiency on semantic segmentation to a new level, as our evaluation results on ADE20K and Cityscapes show. HRViT achieves 50.20% mIoU on ADE20K and 83.16% mIoU on Cityscapes, surpassing state-of-the-art MiT and CSWin backbones with an average of +1.78 mIoU improvement, 28% parameter saving, and 21% FLOPs reduction, demonstrating the potential of HRViT as a strong vision backbone for semantic segmentation.

cs.CV↗

arXiv2021

Deep Embedded K-Means Clustering

Recently, deep clustering methods have gained momentum because of the high representational power of deep neural networks (DNNs) such as autoencoder. The key idea is that representation learning and clustering can reinforce each other: Good representations lead to good clustering while good clustering provides good supervisory signals to representation learning. Critical questions include: 1) How to optimize representation learning and clustering? 2) Should the reconstruction loss of autoencoder be considered always? In this paper, we propose DEKM (for Deep Embedded K-Means) to answer these two questions. Since the embedding space generated by autoencoder may have no obvious cluster structures, we propose to further transform the embedding space to a new space that reveals the cluster-structure information. This is achieved by an orthonormal transformation matrix, which contains the eigenvectors of the within-class scatter matrix of K-means. The eigenvalues indicate the importance of the eigenvectors' contributions to the cluster-structure information in the new space. Our goal is to increase the cluster-structure information. To this end, we discard the decoder and propose a greedy method to optimize the representation. Representation learning and clustering are alternately optimized by DEKM. Experimental results on the real-world datasets demonstrate that DEKM achieves state-of-the-art performance.

cs.LG↗

arXiv2021

Legal Judgment Prediction with Multi-Stage CaseRepresentation Learning in the Real Court Setting

Legal judgment prediction(LJP) is an essential task for legal AI. While prior methods studied on this topic in a pseudo setting by employing the judge-summarized case narrative as the input to predict the judgment, neglecting critical case life-cycle information in real court setting could threaten the case logic representation quality and prediction correctness. In this paper, we introduce a novel challenging dataset from real courtrooms to predict the legal judgment in a reasonably encyclopedic manner by leveraging the genuine input of the case -- plaintiff's claims and court debate data, from which the case's facts are automatically recognized by comprehensively understanding the multi-role dialogues of the court debate, and then learnt to discriminate the claims so as to reach the final judgment through multi-task learning. An extensive set of experiments with a large civil trial data set shows that the proposed model can more accurately characterize the interactions among claims, fact and debate for legal judgment prediction, achieving significant improvements over strong state-of-the-art baselines. Moreover, the user study conducted with real judges and law school students shows the neural predictions can also be interpretable and easily observed, and thus enhancing the trial efficiency and judgment quality.

cs.CL↗

arXiv2021

Capturing Event Argument Interaction via A Bi-Directional Entity-Level Recurrent Decoder

Capturing interactions among event arguments is an essential step towards robust event argument extraction (EAE). However, existing efforts in this direction suffer from two limitations: 1) The argument role type information of contextual entities is mainly utilized as training signals, ignoring the potential merits of directly adopting it as semantically rich input features; 2) The argument-level sequential semantics, which implies the overall distribution pattern of argument roles over an event mention, is not well characterized. To tackle the above two bottlenecks, we formalize EAE as a Seq2Seq-like learning problem for the first time, where a sentence with a specific event trigger is mapped to a sequence of event argument roles. A neural architecture with a novel Bi-directional Entity-level Recurrent Decoder (BERD) is proposed to generate argument roles by incorporating contextual entities' argument role predictions, like a word-by-word text generation process, thereby distinguishing implicit argument distribution patterns within an event more accurately.

cs.CL↗

arXiv2021

Multi-view Inference for Relation Extraction with Uncertain Knowledge

Knowledge graphs (KGs) are widely used to facilitate relation extraction (RE) tasks. While most previous RE methods focus on leveraging deterministic KGs, uncertain KGs, which assign a confidence score for each relation instance, can provide prior probability distributions of relational facts as valuable external knowledge for RE models. This paper proposes to exploit uncertain knowledge to improve relation extraction. Specifically, we introduce ProBase, an uncertain KG that indicates to what extent a target entity belongs to a concept, into our RE architecture. We then design a novel multi-view inference framework to systematically integrate local context and global knowledge across three views: mention-, entity- and concept-view. The experimental results show that our model achieves competitive performances on both sentence- and document-level relation extraction, which verifies the effectiveness of introducing uncertain knowledge and the multi-view inference framework that we design.

cs.CL↗

arXiv2021

Exploiting Method Names to Improve Code Summarization: A Deliberation Multi-Task Learning Approach

Code summaries are brief natural language descriptions of source code pieces. The main purpose of code summarization is to assist developers in understanding code and to reduce documentation workload. In this paper, we design a novel multi-task learning (MTL) approach for code summarization through mining the relationship between method code summaries and method names. More specifically, since a method's name can be considered as a shorter version of its code summary, we first introduce the tasks of generation and informativeness prediction of method names as two auxiliary training objectives for code summarization. A novel two-pass deliberation mechanism is then incorporated into our MTL architecture to generate more consistent intermediate states fed into a summary decoder, especially when informative method names do not exist. To evaluate our deliberation MTL approach, we carried out a large-scale experiment on two existing datasets for Java and Python. The experiment results show that our technique can be easily applied to many state-of-the-art neural models for code summarization and improve their performance. Meanwhile, our approach shows significant superiority when generating summaries for methods with non-informative names.

cs.CL↗

arXiv2021

Improvement of phase sensitivity in SU(1,1) interferometer via a Kerr nonlinear

We propose a theoretical scheme to enhance the phase sensitivity by introducing a Kerr nonlinear phase shift into the traditional SU(1,1) interferometer with a coherent state input and homodyne detection. We investigate the realistic effects of photon losses on phase sensitivity and quantum Fisher information. The results show that compared with the linear phase shift in SU(1,1) interferometer, the Kerr nonlinear case can not only enhance the phase sensitivity and quantum Fisher information, but also significantly suppress the photon losses. We also observe that at the same accessible parameters, internal losses have a greater influence on the phase sensitivity than the external ones. It is interesting that, our scheme shows an obvious advantage of low-cost input resources to obtain higher phase sensitivity and larger quantum Fisher information due to the introduction of nonlinear phase element.

quant-ph↗

arXiv2021

Operator transpose within normal ordering and its applications for quantifying entanglement

Partial transpose is an important operation for quantifying the entanglement, here we study the (partial) transpose of any single (two-mode) operators. Using the Fock-basis expansion, it is found that the transposed operator of an arbitrary operator can be obtained by replacement of a^{†}(a) by a(a^{†}) instead of c-number within normal ordering form. The transpose of displacement operator and Wigner operator are studied, from which the relation of Wigner function, characteristics function and average values such as covariance matrix are constructed between density operator and transposed density operator. These observations can be further extended to multi-mode cases. As applications, the partial transpose of two-mode squeezed operator and the entanglement of two-mode squeezed vacuum through a laser channel are considered.

quant-ph↗

arXiv2020

SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint

Automatic song writing aims to compose a song (lyric and/or melody) by machine, which is an interesting topic in both academia and industry. In automatic song writing, lyric-to-melody generation and melody-to-lyric generation are two important tasks, both of which usually suffer from the following challenges: 1) the paired lyric and melody data are limited, which affects the generation quality of the two tasks, considering a lot of paired training data are needed due to the weak correlation between lyric and melody; 2) Strict alignments are required between lyric and melody, which relies on specific alignment modeling. In this paper, we propose SongMASS to address the above challenges, which leverages masked sequence to sequence (MASS) pre-training and attention based alignment modeling for lyric-to-melody and melody-to-lyric generation. Specifically, 1) we extend the original sentence-level MASS pre-training to song level to better capture long contextual information in music, and use a separate encoder and decoder for each modality (lyric or melody); 2) we leverage sentence-level attention mask and token-level attention constraint during training to enhance the alignment between lyric and melody. During inference, we use a dynamic programming strategy to obtain the alignment between each word/syllable in lyric and note in melody. We pre-train SongMASS on unpaired lyric and melody datasets, and both objective and subjective evaluations demonstrate that SongMASS generates lyric and melody with significantly better quality than the baseline method without pre-training or alignment constraint.

cs.SD↗

arXiv2020

Improving phase estimation using the number-conserving operations

We propose a theoretical scheme to improve the resolution and precision of phase measurement with parity detection in the Mach-Zehnder interferometer by using a nonclassical input state which is generated by applying a number-conserving generalized superposition of products (GSP) operation, (saa^{†}+ta^{†}a)^{m} with s^2+t^2=1, on two-mode squeezed vacuum (TMSV) state. The nonclassical properties of the proposed GSP-TMSV are investigated via average photon number (APN), anti-bunching effect, and degrees of two-mode squeezing. Particularly, our results show that both higher-order m GSP operation and smaller parameter s can increase the total APN, which leads to the improvement of quantum Fisher information. In addition, we also compare the phase measurement precision with and without photon losses between our scheme and the previous photon subtraction/addition schemes. It is found that our scheme, especially for the case of s=0, has the best performance via the enhanced phase resolution and sensitivity when comparing to those previous schemes even in the presence of photon losses. Interestingly, without losses, the standard quantum-noise limit (SQL) can always be surpassed in our our scheme and the Heisenberg limit (HL) can be even achieved when s=0.5,1 with small total APNs. However, in the presence of photon losses, the HL cannot be beaten, but the SQL can still be overcome particularly in the large total APN regimes. Our results here can find important applications in quantum metrology.

quant-ph↗

arXiv2020

Enhancing discrete-modulated continuous-variable measurement-device-independent quantum key distribution via quantum catalysis

The discrete modulation can make up for the shortage of transmission distance in measurement-device-independent continuous-variable quantum key distribution (MDI-CVQKD) that has an unique advantage against all side-channel attacks but also challenging for the further performance improvement. Here we suggest a quantum catalysis (QC) approach for enhancing the performance of the discrete-modulated (DM) MDI-CVQKD in terms of the achievable secret key rate and lengthening the maximal transmission distance. The numerical simulation results show that the QC-based MDI-CVQKD with discrete modulation that involves a zero-photon catalysis (ZPC) operation can not only obtain a higher secret key rate than the original DM protocol, but also contributes to the reasonable increase of the corresponding optimal variance. As for the extreme asymmetric and symmetric cases, the secret key rate and maximal transmission distance of the ZPC-involved DM MDI-CVQKD system can be further improved under the same parameters. This approach enables the system to tolerate lower reconciliation efficiency, which will promote the practical implementations with state-of-art technology.

quant-ph↗

arXiv2020

Incorporating User's Preference into Attributed Graph Clustering

Graph clustering has been studied extensively on both plain graphs and attributed graphs. However, all these methods need to partition the whole graph to find cluster structures. Sometimes, based on domain knowledge, people may have information about a specific target region in the graph and only want to find a single cluster concentrated on this local region. Such a task is called local clustering. In contrast to global clustering, local clustering aims to find only one cluster that is concentrating on the given seed vertex (and also on the designated attributes for attributed graphs). Currently, very few methods can deal with this kind of task. To this end, we propose two quality measures for a local cluster: Graph Unimodality (GU) and Attribute Unimodality (AU). The former measures the homogeneity of the graph structure while the latter measures the homogeneity of the subspace that is composed of the designated attributes. We call their linear combination as Compactness. Further, we propose LOCLU to optimize the Compactness score. The local cluster detected by LOCLU concentrates on the region of interest, provides efficient information flow in the graph and exhibits a unimodal data distribution in the subspace of the designated attributes.

cs.LG↗

arXiv2020

Leveraging Code Generation to Improve Code Retrieval and Summarization via Dual Learning

Code summarization generates brief natural language description given a source code snippet, while code retrieval fetches relevant source code given a natural language query. Since both tasks aim to model the association between natural language and programming language, recent studies have combined these two tasks to improve their performance. However, researchers have yet been able to effectively leverage the intrinsic connection between the two tasks as they train these tasks in a separate or pipeline manner, which means their performance can not be well balanced. In this paper, we propose a novel end-to-end model for the two tasks by introducing an additional code generation task. More specifically, we explicitly exploit the probabilistic correlation between code summarization and code generation with dual learning, and utilize the two encoders for code summarization and code generation to train the code retrieval task via multi-task learning. We have carried out extensive experiments on an existing dataset of SQL and Python, and results show that our model can significantly improve the results of the code retrieval task over the-state-of-art models, as well as achieve competitive performance in terms of BLEU score for the code summarization task.

cs.IR↗

arXiv2020

Tree++: Truncated Tree Based Graph Kernels

Graph-structured data arise ubiquitously in many application domains. A fundamental problem is to quantify their similarities. Graph kernels are often used for this purpose, which decompose graphs into substructures and compare these substructures. However, most of the existing graph kernels do not have the property of scale-adaptivity, i.e., they cannot compare graphs at multiple levels of granularities. Many real-world graphs such as molecules exhibit structure at varying levels of granularities. To tackle this problem, we propose a new graph kernel called Tree++ in this paper. At the heart of Tree++ is a graph kernel called the path-pattern graph kernel. The path-pattern graph kernel first builds a truncated BFS tree rooted at each vertex and then uses paths from the root to every vertex in the truncated BFS tree as features to represent graphs. The path-pattern graph kernel can only capture graph similarity at fine granularities. In order to capture graph similarity at coarse granularities, we incorporate a new concept called super path into it. The super path contains truncated BFS trees rooted at the vertices in a path. Our evaluation on a variety of real-world graphs demonstrates that Tree++ achieves the best classification accuracy compared with previous graph kernels.

cs.LG↗

arXiv2019

Continuous-variable measurement-device-independent quantum key distribution via quantum catalysis

The continuous-variable measurement-device-independent quantum key distribution (CV-MDI-QKD) is a promising candidate for the immunity to side-channel attacks, but unfortunately seems to face the limitation of transmission distance in contrast to discrete-variable (DV) counterpart. In this paper, we suggest a method of improving the performance of CV-MDI-QKD involving the achievable secret key rate and transmission distance by using zero-photon catalysis (ZPC), which is indeed a noiseless attenuation process. The numerical stimulation results show that the transmission distance of ZPC-based CV-MDI-QKD under the extreme asymmetric case is better than that of the original protocol. Attractively, in contrast to the previous single-photon subtraction (SPS)-based CV-MDI-QKD, the proposed scheme enables a higher secret key rate and a longer transmission distance. In particular, the ZPC-based CV-MDI-QKD can tolerate more imperfections of detectors than both the original protocol and the SPS-based CV-MDI-QKD.

quant-ph↗