Assoc Prof Dr. Chuanzhong Wu | Deep Metric Learning | Outstanding Scientist Award

Posted on by ScienceFather

Assoc. Prof. Dr. Chuanzhong Wu | Deep Metric Learning | Outstanding Scientist Award

Chuanzhong Wu at Shanghai International Studies University, China

Profiles

Scopus

🎓 Early Academic Pursuits

Assoc. Prof. Dr. Chuanzhong Wu embarked on his academic journey with a Bachelor’s degree in Physical Education from Wuhan Institute of Physical Education in 2005. His passion for sports education and training led him to pursue a Master’s degree in Sports Education & Training Science at the same institution, which he completed in 2008. Driven by a commitment to advancing research in sports humanities, he earned his Ph.D. in Sports Humanities and Social Sciences from the National University of Physical Education and Sport of Ukraine in September 2023. His doctoral studies focused on the intersection of sports education and social sciences, under the supervision of Prof. Korobeynikava Lesia.

🏢 Professional Endeavors

Assoc Prof Dr. Wu began his teaching career in 2008 as a Teaching Assistant at Huaihai Institute of Technology. Over the years, he progressed through various academic ranks, becoming a Lecturer in 2010 and later achieving the title of Associate Professor in 2018. Currently, he serves as an Associate Professor at Jiangsu Ocean University, where he holds the position of Section Chief in the Department of Sports. His dedication to academia and sports training has earned him recognition as a key figure in sports education and talent development.

🔬 Contributions and Research Focus

Assoc Prof Dr. Wu’s research is centered on Sports Education and Training Science, where he explores innovative training methodologies, physical conditioning, and the social dimensions of sports. His work has significantly contributed to enhancing the understanding of sports culture, performance analysis, and athletic training strategies. Through extensive research and publications, he has examined topics such as the integration of school and community sports culture and the relationship between competitive sports origin theories and human demand for multi-level sports development.

🌍 Impact and Influence

As a recognized researcher in the field,Assoc Prof Dr. Wu has made substantial contributions to the academic community. His work has been honored on multiple occasions, including First Prize at the European Youth Olympic Scientific Paper Conference (2020) and Second Prize at the 2020 Tokyo Olympic Games Scientific Paper Conference. His research findings have not only influenced sports training methodologies but also contributed to policy recommendations and curriculum development in higher education institutions.

📚 Academic Cites and Recognitions

Assoc Prof Dr. Wu’s academic excellence has been acknowledged through various city and provincial-level awards. In 2021, he was selected for Lianyungang City’s “521 High-Level Talent Training Program” as a Third-Tier Scholar. His research papers have received accolades in prestigious competitions, including:

  • Second Prize in the 13th National Student Sports Conference Scientific Paper Competition (2017)
  • Second Prize in the National College Student Work Excellent Academic Achievement Award (2012)
  • Recognition as an Outstanding Instructor for University Students’ Summer Social Practice Program (2012)

💻 Technical Skills

Assoc Prof Dr. Wu has extensive expertise in sports performance analysis, physical education methodologies, emergency rescue training, and sports research analytics. His technical skills include quantitative research methods, data-driven training assessments, and interdisciplinary sports education approaches. He is also proficient in designing and implementing sports training programs that bridge traditional education and modern technological applications.

🎓 Teaching Experience and Student Engagement

Throughout his teaching career,Assoc Prof Dr. Wu has been widely recognized for his student-centered approach and commitment to academic excellence. In 2017, he was voted the “Most Beloved Teacher” by students at Jiangsu Ocean University. His dedication to mentorship has earned him multiple awards as an “Outstanding Class Advisor” over consecutive years. His courses emphasize scientific training techniques, sports psychology, and athletic development, inspiring students to pursue excellence in sports and academia.

🌟 Legacy and Future Contributions

Assoc Prof Dr. Wu’s impact in the field of sports education and training science continues to grow. As a dedicated researcher and educator, he strives to bridge the gap between theoretical research and practical sports applications. His future contributions aim to enhance global sports training methodologies, promote interdisciplinary research, and develop next-generation athletes through innovative educational frameworks. With a strong foundation in research, teaching, and leadership, Dr. Wu remains committed to shaping the future of sports education and training science on both a national and international scale.

 

Publications

Infrared Thermal Radiation and Deep Learning Algorithms for Evaluating the Warm-Up Effect of Sports Training: Thermal Imaging Monitoring Model

  • Author: Y. Liu, Yumeng; Y. Li, Yunlong; D. Liang, Danqing; C. Li, Cheng; C. Wu, Chuanzhong
    Journal: Thermal Science and Engineering Progress
    Year: 2025

Dr. Na Yi | Deep Metric Learning | Best Researcher Award

Posted on by ScienceFather

Dr. Na Yi | Deep Metric Learning | Best Researcher Award

Doctorate at Heilongjiang University of Science and Technology, China

Profiles

Scopus

Orcid

Academic Background

Dr. Na Yi, born in June 1997 in Acheng, Harbin, is an Associate Professor and a committed member of the Communist Party of China. With a strong academic foundation in Electrical Engineering and Automation, she has quickly risen as a prominent figure in the field of Petroleum and Natural Gas Engineering.

Education

Dr. Na Yi graduated with a degree in Electrical Engineering and Automation from Northeast Petroleum University in 2019. She was subsequently recommended for a doctoral program in Petroleum and Natural Gas Engineering, during which she also studied at Southeast University, earning her doctorate in 2024.

Professional Experience

Throughout her career, Dr. Na Yi has published over 20 research papers in esteemed journals, with 10 SCI-indexed and 5 EI-indexed papers, including highly cited and hot papers. She holds 6 national patents and has participated in 5 significant scientific research projects. Her achievements have earned her more than 10 national and provincial awards.

Research Interests

Dr. Na Yi’s research interests lie in Petroleum Engineering, with a focus on sustainable energy, power systems, and technological innovation. She is an active reviewer for multiple international and Chinese academic journals and has been invited to present her research at several international and domestic conferences.

 Publications

A multi-stage low-cost false data injection attack method for power CPS

  • Authors: Yi, N., Xu, J., Chen, Y., Pan, F.
  • Journal: Zhejiang Electric Power
  • Year: 2023
A New Distributed Power Supply for Distribution Network Considering SOP Access
  • Authors: Peng, C., Xu, J., Zhao, S., Yi, N.
  • Year: 2023
Multi-stage coordinated cyber-physical topology attack method based on deep reinforcement learning
  • Authors: Yi, N., Xu, J., Chen, Y., Sun, D.
  • Journal: Electric Power Engineering Technology
  • Year: 2023
A multi-stage game model for the false data injection attack from attacker’s perspective
  • Authors: Yi, N., Wang, Q., Yan, L., Tang, Y., Xu, J.
  • Journal: Sustainable Energy, Grids and Networks
  • Year: 2021
Insulator Self-Explosion Defect Detection Based on Hierarchical Multi-Task Deep Learning
  • Authors: Xu, J., Huang, L., Yan, L., Yi, N.
  • Journal: Diangong Jishu Xuebao/Transactions of China Electrotechnical Society
  • Year: 2021

Deep Metric Learning

Posted on by ScienceFather

Introduction of Deep Metric Learning

Introduction: Deep Metric Learning is a specialized field within machine learning and computer vision that focuses on training deep neural networks to learn similarity metrics between data points. It aims to discover meaningful representations of data that enable the computation of distances or similarities between samples, which can be useful in various applications, such as image retrieval, face recognition, and recommendation systems. Deep Metric Learning has gained significant attention due to its potential to improve the performance of similarity-based tasks.

Subtopics in Deep Metric Learning:

  1. Siamese Networks: Siamese networks are a foundational architecture in deep metric learning. Researchers in this subfield explore variations and improvements to Siamese networks, which consist of two identical subnetworks that learn to minimize the distance between similar samples and maximize the distance between dissimilar ones.
  2. Triplet Networks: Triplet networks are designed to learn embeddings where the distance between anchor-positive pairs is minimized and the distance between anchor-negative pairs is maximized. Research focuses on triplet loss variations and effective sampling strategies to improve training stability and convergence.
  3. Margin-Based Losses: Margin-based loss functions, like contrastive loss and triplet margin loss, play a key role in deep metric learning. Researchers work on designing and adapting margin-based loss functions to different tasks and datasets to enhance the discriminative power of learned embeddings.
  4. Hard and Semi-Hard Negative Mining: Mining hard or semi-hard negative samples during training is critical for the success of deep metric learning. This subtopic explores strategies to efficiently select challenging negative samples that help improve model performance.
  5. Multi-Modal Metric Learning: Extending deep metric learning to handle data from multiple modalities, such as text and images, to enable cross-modal similarity calculations, which have applications in recommendation systems and content-based retrieval.

Deep Metric Learning research is essential for creating powerful models capable of understanding and leveraging the inherent similarities and differences in data. These subtopics reflect the ongoing efforts to refine techniques and develop robust deep metric learning models for diverse real-world applications.