Lingfei Deng

Huazhong University of Science and Technology

Papers

Total Citations

331

H-Index

About

Lingfei Deng has made transformative contributions to the field of transfer learning, with a particular focus on understanding and mitigating the phenomenon of negative transfer—where knowledge from source domains inadvertently degrades performance in a target domain. Their seminal survey, “A Survey on Negative Transfer” (2022), has already garnered over 331 citations, establishing it as a foundational reference for researchers grappling with the pitfalls of domain adaptation. Deng’s work systematically categorizes the causes of negative transfer, from domain dissimilarity to label-space mismatches, and proposes novel frameworks for detecting and avoiding such harmful interference. By illuminating these challenges, Deng has not only advanced theoretical understanding but also provided practical guidelines for building more robust transfer learning systems in data-scarce environments—critical for applications like medical imaging and privacy-preserving analytics. Their research bridges the gap between theory and application, offering clear taxonomies and actionable solutions that have shaped subsequent work in the field. Deng’s contributions are essential reading for any student or researcher seeking to harness the full potential of transfer learning while avoiding its hidden traps.