Yi Zhang

Yi Zhang

Auto-generated profile, based on publicly available information

Yi Zhang

State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
1 July 2019
Verified
Conference Paper
Open access
YZYi Zhang
Yi Zhang
CZCheng Zeng
Cheng Zeng
HCHao Cheng
Hao Cheng
CWChongjun Wang
Chongjun Wang
LZLei Zhang
Lei Zhang
With the emergence of diverse data collection techniques, objects in real applications can be represented as multi-modal features. What's more, objects may have multiple semantic meanings. Multi-modal and Multi-label [1] (MMML) problem becomes a universal phenomenon. The quality of data collected from different channels are inconsistent and some of them may not benefit for prediction. In real life, not all the modalities are needed for prediction. As a result, we propose a novel instance-oriented Multi-modal Classifier Chains (MCC) algorithm for MMML problem, which can make convince prediction with partial modalities. MCC extracts different modalities for different instances in the testing phase. Extensive experiments are performed on one real-world herbs dataset and two public datasets to validate our proposed algorithm, which reveals that it may be better to extract many instead of all of the modalities at hand.

22 October 2021
Verified
Conference Paper
Closed access
LPLing Pang
Ling Pang
LWLulu Wang
Lulu Wang
YZYi Zhang
Yi Zhang
HLHu Li
Hu Li
With the development of deep neural network techniques, an explosion of neural network structures are designed and gained great success in the field of image-based SAR automatic target recognition. Recent studies have shown that machine learning networks are vulnerable to adversarial attacks, which may cause severe security issue. In this paper, a convolution neural network (CNN) for SAR target classification is designed and then attacked by adversarial examples using fast gradient sign method (FGSM) and basic iteration method (BIM). Experiments shown that although the classification accuracy of the model is quite high for real SAR image dataset, elaborately designed tiny perturbation on the data will magnificently reduce the classification performance.

11 January 2022 in IEEE MultiMedia
Verified
Journal Article
Closed access
YZYi Zhang
Yi Zhang
JSJundong Shen
Jundong Shen
CYCheng Yu
Cheng Yu
CWChongjun Wang
Chongjun Wang
Multiview multilabel (MVML) learning deals with objects with diverse feature vectors and rich semantics. Existing methods are built on a shared latent space among multiple views. However, they do not well capture semantic consistency and view interactions. What’s more, they neglect different contributions of each view to multilabel learning. To address these issues, a novel Personalized RElation rEfinement Network (PREEN) is proposed to sufficiently exploit associated relationships. First, common and specific information is learned in an adversarial way to prevent them from interfering with each other. Then, we adapt the standard transformer to capture cross-view interactions. Similarly, we design label-specific transformers to model label-view dependence, which associates each label with relevant views separately. Eventually, we develop an interlabel attention mechanism to exploit label correlations and refine complementary information from other labels dynamically. Extensive experiments on one real and seven public MVML datasets validate the effectiveness of our proposed PREEN.

2 December 2019 in arXiv
Verified
Preprint
Open access
PDF file
HKHiroaki Kaido
Hiroaki Kaido
YZYi Zhang
Yi Zhang
This study develops a framework for testing hypotheses on structural parameters in incomplete models. Such models make set-valued predictions and hence do not generally yield a unique likelihood function. The model structure, however, allows us to construct tests based on the least favorable pairs of likelihoods using the theory of Huber and Strassen (1973). We develop tests robust to model incompleteness that possess certain optimality properties. We also show that sharp identifying restrictions play a role in constructing such tests in a computationally tractable manner. A framework for analyzing the local asymptotic power of the tests is developed by embedding the least favorable pairs into a model that allows local approximations under the limits of experiments argument. Examples of the hypotheses we consider include those on the presence of strategic interaction effects in discrete games of complete information. Monte Carlo experiments demonstrate the robust performance of the proposed tests.

14 September 2022 in Intelligent Data Analysis: An International Journal
Verified
Journal Article
Closed access
PDF file
YZYi Zhang
Yi Zhang
ZZZhecheng Zhang
Zhecheng Zhang
MCMing-Yuan Chen
Ming-Yuan Chen
HLHeng-yang Lu
Heng-yang Lu
LZLei Zhang
Lei Zhang
CWChongjun Wang
Chongjun Wang
Exploiting label correlation is crucially important in multi-label learning, where each instance is associated with multiple labels simultaneously. Multi-label learning is more complex than single-label learning for that the labels tend to be correlated. Traditional multi-label learning algorithms learn independent classifiers for each label and employ ranking or threshold on the classification results. Most existing methods take label correlation as prior knowledge, which have worked well, but they failed to make full use of label dependency. As a result, the real relationship among labels may not be correctly characterized and the final prediction is not explicitly correlated. To address these problems, we propose a novel high-order multi-label learning algorithm of Label collAboration based Multi-laBel learning (LAMB). With regard to each label, LAMB utilizes collaboration between its own prediction and the prediction of other labels. Extensive experiments on various datasets demonstrate that our proposed LAMB algorithm achieves superior performance over existing state-of-the-art algorithms. In addition, one real-world dataset of channelrhodopsins chimeras is assessed, which would be of great value as pre-screen for membrane proteins function.

5 July 2021
Verified
Conference Paper
Closed access
YZYi Zhang
Yi Zhang
JSJundong Shen
Jundong Shen
YCYu Cheng
Yu Cheng
CWChongjun Wang
Chongjun Wang
In Multi-View Multi-Label (MVML) learning, each instance is represented with several feature vectors and associated with multiple labels simultaneously. Existing models are mainly built on a shared latent space among multiple views and tend to select features that are distinguishable for all labels. However, they ignore the distinctive characteristic of each view and fail to extract the discriminative information related to corresponding labels. To address the above challenges, we proposed a novel approach named Label-specific Alignment with Adversarial Multi-view Representation (LAAMR). For one thing, shared and private information is extracted with adversarial training and orthogonal constraint, which ensures complementarity and consistency among multiple views. For another, label semantic information is well captured with multi-head attention, and then utilized as the guidance to exploit label-view dependence with alignment attention, resulting in label-specific representations. Experiments on various MVML datasets validate the effectiveness of LAAMR.