Zehao Liu

Zehao Liu

Auto-generated profile, based on publicly available information

Zehao Liu

College of Information Sciences and Technology, Pennsylvania State University, University Park, USA

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
1 January 2020 in The American Journal of Chinese Medicine
Verified
Journal Article
Closed access
PDF file
YXYong-Yue Xu
Yong-Yue Xu
DWDong-Mei Wang
Dong-Mei Wang
HLHua-Sheng Liang
Hua-Sheng Liang
ZLZe-Hao Liu
Ze-Hao Liu
JLJun-Xia Li
Jun-Xia Li
CLChuanjian Lu
Chuanjian Lu

9 November 2022 in ACM Transactions on Intelligent Systems and Technology
Verified
Journal Article
Open access
WYWenlu Yang
Wenlu Yang
HWHong jun Wang
Hong jun Wang
YZYinghui Zhang
Yinghui Zhang
ZLZehao Liu
Zehao Liu
TLTianrui Li
Tianrui Li
Representation learning based on autoencoders has received great concern for its potential ability to capture valuable latent information. Conventional autoencoders pursue minimal reconstruction error, but in most machine learning tasks such as classification and clustering, the discrimination of feature representation is also important. To address this limitation, an enhanced self-supervised discriminative fuzzy autoencoder (FAE) is innovatively proposed, which focuses on exploring information within data to guide the unsupervised training process and enhancing feature discrimination in a self-supervised manner. In FAE, fuzzy membership is applied to provide a means of self-supervised, which allows FAE can not only utilize AE’s outstanding representation learning capabilities but can also transform the original data into another space with improved discrimination. First, the objective function corresponding to FAE is proposed by reconstruction loss and clustering oriented loss simultaneously. Subsequently, Mini-Batch Gradient Descent is applied to infer the objective function and the detailed process is illustrated step by step. Finally, empirical studies on clustering tasks have demonstrated the superiority of FAE over the state of the art.

12 February 2025
Verified
Conference Paper
Open access
PKPriya C. Kumar
Priya C. Kumar
JSJeffrey Samuel Schulman
Jeffrey Samuel Schulman
FAFatimah Albargi
Fatimah Albargi
SBSree Bhattacharyya
Sree Bhattacharyya
HDHongyi Dong
Hongyi Dong
ZLZehao Liu
Zehao Liu
The process of learning research methods is complex, often propelling students to shift their worldviews as they work to understand concepts and develop skills. We argue that a pedagogy of ungrading, which eschews points-based grading for feedback-based assessment, is well suited to helping students shift their worldviews because it centers engagement over measuring outcomes. We-a faculty instructor and five PhD students-present this experience report detailing the design and implementation of ungrading in a graduate qualitative research methods course in an informatics PhD program. We explain the course design and present self-reflections that illustrate how a pedagogy of ungrading facilitated meaningful learning experiences among students whose research interests span the epistemological and methodological spectrum.

1 August 2015
Verified
Conference Paper
Closed access
LZLiu Zehao
Liu Zehao
QHQian Hanting
Qian Hanting
WCWei Chenjia
Wei Chenjia
Recently, a class of online blind extraction algorithms in the framework of Kalman filtering have been developed for wireless sensor networks applications. But their performance is poor when additive noise is not neglectable. To address this issue, a new blind extraction method is proposed. This method novelly adopts two unscented Kalman filters alternately estimate the extraction vector and the source signal. Thereby, an optimal estimation of the source signal in sense of minimum mean square error is obtained. Simulation result indicates that the proposed algorithm can effectively extract target signal, and outperforms the existing unscented Kalman approach.