Dr. Zexiao Liang has accumulated valuable professional experience as a postdoctoral assistant researcher at the School of Integrated Circuits, Guangdong University of Technology since July 2022. His work primarily focuses on the research and design of machine learning algorithms, particularly in the integration of multi-information fusion and clustering analysis. Dr. Liang has developed algorithms to address specific domain challenges, such as predicting the effects of multi-drug interactions and conducting reliability analyses for low-quality chip images. His role also involves guiding students, which has led to the publication of numerous scientific papers and contributions to various research projects.
Research Interests:
Dr. Zexiao Liang’s research interests lie in the field of multi-information fusion machine learning algorithms, including the integration of multiple transformation domain information and the development of multi-modal, multi-view learning techniques. He is particularly interested in clustering analysis and the design and application of machine learning algorithms to address domain-specific challenges. His work encompasses algorithms for predicting the effects of multi-drug interactions and reliability analysis for low-quality chip images. Dr. Liang is dedicated to advancing machine learning methodologies and their practical applications in solving complex problems in various domains.
Skills:
Dr. Zexiao Liang specializes in advanced machine learning and data analysis techniques, focusing on the integration of multiple transformation domain information, multi-modal and multi-view learning, and clustering analysis. His expertise also includes designing algorithms to address domain-specific challenges, such as predicting multi-drug interactions and analyzing low-quality chip images for reliability. He is proficient in feature fusion strategies and parameter optimization, as well as dimensionality reduction techniques for manifold-based semi-supervised classification.
Publications: