Dehui Du | Computer Science | Innovative Research Award

Published on by Top Teachers

Innovative Research Award

Dehui Du
East China Normal Universty, China

Dehui Du
Affiliation East China Normal Universty
Country China
Scopus ID 14044898400
Documents 68
Citations 504
h-index 11
Subject Area Computer Science
Event Top Teachers Awards

The Innovative Research Award recognizes scholars whose research activities demonstrate originality, methodological rigor, and measurable contributions to the advancement of scientific knowledge. Dehui Du of East China Normal Universty has established a research profile in computer science through investigations in causal inference, explainable artificial intelligence, reinforcement learning, large language models, autonomous systems, and rare event detection. His publication record, citation performance, and participation in internationally recognized conferences indicate sustained engagement with contemporary research challenges and emerging computational methodologies.[1]

Abstract

Dehui Du’s research focuses on the intersection of machine learning, causal reasoning, explainable artificial intelligence, and intelligent systems. His scholarly output addresses practical and theoretical problems associated with reinforcement learning, counterfactual analysis, autonomous driving, and large language models. Through conference publications and collaborative research efforts, he has contributed to the development of computational frameworks designed to improve transparency, reliability, and performance in artificial intelligence systems.[2]

Keywords

Artificial Intelligence, Computer Science, Reinforcement Learning, Causal Inference, Explainable AI, Large Language Models, Counterfactual Analysis, Autonomous Driving.

Introduction

Recent advances in artificial intelligence increasingly require interpretable, reliable, and data-efficient learning systems. Researchers working at the intersection of machine learning and causal reasoning play an important role in addressing these challenges. Dehui Du’s work reflects this direction by integrating explainability, counterfactual reasoning, and advanced learning architectures into practical computational frameworks that support decision-making and predictive performance.[3]

Research Profile

With 68 indexed publications, 504 citations, and an h-index of 11, Dehui Du has developed a scholarly profile characterized by interdisciplinary research across machine learning and intelligent computing. His collaborations span topics including causal inference, experience replay methods, language model reasoning, autonomous systems, and counterfactual identifiability. These areas are increasingly relevant to both academic research and industrial applications.[1]

Research Contributions

  • Development of explainable reinforcement learning approaches supported by causal inference.
  • Advancement of counterfactual generation techniques for rare event detection.
  • Research on preference-guided reverse reasoning for large language models.
  • Theoretical investigations into exogenous isomorphism and counterfactual identifiability.
  • Contributions to imitation learning frameworks for autonomous driving systems.

Publications

  1. Enhancing Rare Event Detection via Counterfactual Generation with Exogenous Variables.
  2. ERCI: An Explainable Experience Replay Approach with Causal Inference for Deep Reinforcement Learning.
  3. Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up.
  4. Exogenous Isomorphism for Counterfactual Identifiability.
  5. Multi-Task Invariant Representation Imitation Learning for Autonomous Driving.

Research Impact

The research output of Dehui Du demonstrates influence across multiple areas of artificial intelligence. His publications appear in recognized venues such as WWW, AAAI, ACL, ICML, and ICRA, reflecting engagement with leading scholarly communities. The combination of theoretical and applied research contributes to improved interpretability, reliability, and effectiveness of machine learning systems in real-world environments.[4]

Award Suitability

Dehui Du’s academic accomplishments align with the objectives of the Innovative Research Award. His work addresses contemporary challenges in artificial intelligence through innovative methodologies and interdisciplinary perspectives. The quality of publication venues, measurable citation indicators, and contributions to explainable and trustworthy AI collectively support consideration for recognition within the Top Teachers Awards framework.[5]

Conclusion

The scholarly record of Dehui Du reflects sustained contributions to computer science research, particularly in machine learning, causal inference, and intelligent systems. Through publications, collaborations, and methodological innovations, he has contributed to the advancement of explainable and reliable artificial intelligence technologies. These achievements provide a strong foundation for recognition through the Innovative Research Award.

References

  1. Elsevier. (n.d.). Scopus author details: Dehui Du, Author ID 14044898400. Scopus.
    https://www.scopus.com/authid/detail.uri?authorId=14044898400
  2. Du, D., Tian, L., Chen, Y., Li, Y., & Li, Y. (2025). ERCI: An Explainable Experience Replay Approach with Causal Inference for Deep Reinforcement Learning.
  3. Yuan, J., Du, D., Zhang, H., Di, Z., & Naseem, U. (2025). Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up.
  4. Chen, Y., & Du, D. (2025). Exogenous Isomorphism for Counterfactual Identifiability.
  5. Peng, J., Yu, X., Wang, J., Tian, L., & Du, D. (2025). Multi-Task Invariant Representation Imitation Learning for Autonomous Driving.
  6. Tian, L., Du, D., & Chen, Y. (2026). Enhancing Rare Event Detection via Counterfactual Generation with Exogenous Variables.