Duan Junhua

Papers

Total Citations

H-Index

About

Duan Junhua is a researcher focused on advancing artificial intelligence and robotics through innovative knowledge representation and reinforcement learning techniques. His primary research areas include cognitive modeling, heterogeneous data integration, and robot learning algorithms. In his most impactful work, "A Cognition Knowledge Representation Model Based on Multidimensional Heterogeneous Data" (2020), Duan addressed a critical challenge in industrial Internet environments: the need to capture the diversity, semantics, hierarchy, and relevance of environmental information. Unlike existing methods that overemphasize simple concepts and relationships, his model provides a more comprehensive framework for representing complex, multidimensional data, earning 16 citations and establishing a foundation for smarter autonomous systems. Duan also contributed to reinforcement learning with "An Improved Tentative Q Learning Algorithm for Robot Learning" (2018), which explores more efficient exploration strategies for robotic agents. While this work has garnered 2 citations, it reflects his ongoing commitment to enhancing machine learning in practical robotics. Duan’s research bridges the gap between theoretical cognition models and real-world industrial applications, offering valuable insights for students and researchers working on intelligent systems, data fusion, and autonomous decision-making.