Zenglin Xu

Papers

Total Citations

H-Index

About

Zenglin Xu is a leading researcher at the intersection of machine learning, tensor methods, and neural network theory. His most influential work bridges the gap between tensor networks and deep learning, as exemplified by his highly cited survey "Tensor Networks Meet Neural Networks: A Survey and Future Perspectives" (2023, 14 citations), which systematically explores how tensor decompositions can mitigate the curse of dimensionality in large-scale models. Xu’s major contributions include developing novel tensor-based architectures that reduce exponential computational complexity to polynomial scales, enabling more efficient training of deep neural networks. His research has profound implications for high-dimensional data analysis, quantum machine learning, and scalable AI systems. Beyond this landmark survey, Xu has published extensively on optimization algorithms, probabilistic graphical models, and large-scale learning systems, earning recognition for advancing both theoretical foundations and practical implementations. With a growing citation impact, his work continues to shape how researchers integrate tensor networks with modern neural architectures, offering elegant solutions to fundamental challenges in computational efficiency and model expressivity.