Yue Yu

Yue Yu

Auto-generated profile, based on publicly available information

Yue Yu

Pengcheng Laboratory, Shenzhen, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 April 2023 in Nature Machine Intelligence
Verified
Journal Article
Open access
PDF file
XWXiaoqi Wang
Xiaoqi Wang
YCYingjie Cheng
Yingjie Cheng
YYYaning Yang
Yaning Yang
YYYue Yu
Yue Yu
FLFei Li
Fei Li
XZXiaoyu Zhang
Xiaoyu Zhang
Self-supervised representation learning (SSL) on biomedical networks provides new opportunities for drug discovery; however, effectively combining multiple SSL models is still challenging and has been rarely explored. We therefore propose multitask joint strategies of SSL on biomedical networks for drug discovery, named MSSL2drug. We design six basic SSL tasks that are inspired by the knowledge of various modalities, inlcuding structures, semantics and attributes in heterogeneous biomedical networks. Importantly, fifteen combinations of multiple tasks are evaluated using a graph-attention-based multitask adversarial learning framework in two drug discovery scenarios. The results suggest two important findings: (1) combinations of multimodal tasks achieve better performance than other multitask joint models; (2) the local–global combination models yield higher performance than random two-task combinations when there are the same number of modalities. We thus conjecture that the multimodal and local–global combination strategies can be treated as the guideline of multitask SSL for drug discovery.

18 July 2016 in Science China Information Sciences
Verified
Journal Article
Closed access
PDF file
YYYue Yu
Yue Yu
GYGang Yin
Gang Yin
TWTao Wang
Tao Wang
CYCheng Yang
Cheng Yang
HWHuaimin Wang
Huaimin Wang

1 November 2023 in IEEE Communications Surveys & Tutorials
Verified
Journal Article
Closed access
STShujiong Tang
Shujiong Tang
YYYue Yu
Yue Yu
HWHui Wang
Hui Wang
GWGuiliang Wang
Guiliang Wang
WCWuhui Chen
Wuhui Chen
WGWen Gao
Wen Gao
The computing demand for massive applications has led to the ubiquitous deployment of computing power. This trend results in the urgent need for higher-level computing resource scheduling services. The Computing and Network Convergence (CNC), a new type of infrastructure, has become a hot topic. To realize the visions of CNC, such as computing-network integration, ubiquitous collaboration, latency-free, and ready-to-use, an intelligent scheduling strategy for CNC should integrate and collaborate with the network. However, the Computing and Network Convergence is built on the cloud, edge, and endless terminals, making the scheduling problem more difficult due to its wide-area requests, available flexibility arrangements, interconnections, and resource adaptations. In view of this, in this survey, we comprehensively review the literature on scheduling in various scenarios. We cover the scheduling problem of Computing and Network Convergence from heterogeneous resources, multiple-objective optimization, and diverse tasks. Possible explanations and implications are discussed. Finally, we point out important challenges for future work.

5 April 2022 in IEEE Transactions on Software Engineering
Verified
Journal Article
Closed access
XZXunhui Zhang
Xunhui Zhang
YYYue Yu
Yue Yu
GGGeorgios Gousios
Georgios Gousios
ARAyushi Rastogi
Ayushi Rastogi
Context: The pull-based development model is widely used in open source projects, leading to the emergence of trends in distributed software development. One aspect that has garnered significant attention concerning pull request decisions is the identification of explanatory factors. Objective: This study builds on a decade of research on pull request decisions and provides further insights. We empirically investigate how factors influence pull request decisions and the scenarios that change the influence of such factors. Method: We identify factors influencing pull request decisions on GitHub through a systematic literature review and infer them by mining archival data. We collect a total of 3,347,937 pull requests with 95 features from 11,230 diverse projects on GitHub. Using these data, we explore the relations among the factors and build mixed effects logistic regression models to empirically explain pull request decisions. Results: Our study shows that a small number of factors explain pull request decisions, with that concerning whether the integrator is the same as or different from the submitter being the most important factor. We also note that the influence of factors on pull request decisions change with a change in context; e.g., the area hotness of pull request is important only in the early stage of project development, however it becomes unimportant for pull request decisions as projects become mature.

30 March 2023 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
CZChen Zeng
Chen Zeng
YYYue Yu
Yue Yu
SLShanshan Li
Shanshan Li
XXXin Xia
Xin Xia
ZWZhiming Wang
Zhiming Wang
XLXiangke Liao
Xiangke Liao
With the rapid increase of public code repositories, developers maintain a great desire to retrieve precise code snippets by using natural language. Despite existing deep learning-based approaches that provide end-to-end solutions (i.e., accept natural language as queries and show related code fragments), the performance of code search in the large-scale repositories is still low in accuracy because of the code representation (e.g., AST) and modeling (e.g., directly fusing features in the attention stage). In this paper, we propose a novel learnable de ep G raph for C ode S earch (called deGraphCS ) to transfer source code into variable-based flow graphs based on an intermediate representation technique, which can model code semantics more precisely than directly processing the code as text or using the syntax tree representation. Furthermore, we propose a graph optimization mechanism to refine the code representation and apply an improved gated graph neural network to model variable-based flow graphs. To evaluate the effectiveness of deGraphCS , we collect a large-scale dataset from GitHub containing 41,152 code snippets written in the C language and reproduce several typical deep code search methods for comparison. The experimental results show that deGraphCS can achieve state-of-the-art performance and accurately retrieve code snippets satisfying the needs of the users.

14 June 2024 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
YYYanwu Yang
Yanwu Yang
CYChenfei Ye
Chenfei Ye
GSGuinan Su
Guinan Su
ZZZiyao Zhang
Ziyao Zhang
ZCZhikai Chang
Zhikai Chang
TMTing Ma
Ting Ma
Foundation models pretrained on large-scale datasets via self-supervised learning demonstrate exceptional versatility across various tasks. Due to the heterogeneity and hard-to-collect medical data, this approach is especially beneficial for medical image analysis and neuroscience research, as it streamlines broad downstream tasks without the need for numerous costly annotations. However, there has been limited investigation into brain network foundation models, limiting their adaptability and generalizability for broad neuroscience studies. In this study, we aim to bridge this gap. In particular, (1) we curated a comprehensive dataset by collating images from 30 datasets, which comprises 70,781 samples of 46,686 participants. Moreover, we introduce pseudo-functional connectivity (pFC) to further generates millions of augmented brain networks by randomly dropping certain timepoints of the BOLD signal. (2) We propose the BrainMass framework for brain network self-supervised learning via mask modeling and feature alignment. BrainMass employs Mask-ROI Modeling (MRM) to bolster intra-network dependencies and regional specificity. Furthermore, Latent Representation Alignment (LRA) module is utilized to regularize augmented brain networks of the same participant with similar topological properties to yield similar latent representations by aligning their latent embeddings. Extensive experiments on eight internal tasks and seven external brain disorder diagnosis tasks show BrainMass’s superior performance, highlighting its significant generalizability and adaptability. Nonetheless, BrainMass demonstrates powerful few/zero-shot learning abilities and exhibits meaningful interpretation to various diseases, showcasing its potential use for clinical applications.