Longwei Wang Joins the University of South Dakota
Dr. Longwei Wang has joined the University of South Dakota as an Assistant Professor. Prior to joining the University of South Dakota, Dr. Wang was an Assistant Professor at Nanjing University of Information Science & Technology, China.
In summer 2022, Longwei Wang successfully defended a doctoral dissertation titled Behavior Analysis and Enhancement of Robustness for Deep Neural Networks, co-advised by Dr. Xiao Qin and Dr. Ashish Gupta. Dr. Wang received the Ph.D. degree in Computer Science and Software Engineering from Auburn University in 2022.
Longwei Wang received the B.E. degree in Electrical Engineering from Southeast University, China, in 2009, the M.E. degree in Information Engineering from Zhejiang University, China, in 2012, and the Ph.D. degree in Electrical Engineering from the University of Texas at Arlington in 2017. From 2018 to 2020, Dr. Wang was a Presidential Graduate Research Fellow in the Department of Computer Science and Software Engineering at Auburn University. Dr. Wang's research interests include Explainable and Trustworthy AI, Robust Deep Learning, Adversarial Robustness, and Machine Learning for Wireless Networks.
During the doctoral study at Auburn, Dr. Wang worked on several projects related to understanding and improving the robustness of deep neural networks. In the first part of the dissertation research, Dr. Wang leveraged an information-theoretical method to study the neuron activation behavior of well-trained classification models, and discovered that the randomness of neuron activation patterns is reduced with the depth of the fully connected layers. In the second part, Dr. Wang incorporated a diversity of symmetries, including rotation and scaling, into convolutional neural networks, showing that the symmetry enforcement makes a model generalize to shifted, rotated, or scaled inputs and substantially strengthens its adversarial robustness. The third part of the dissertation developed a two-stage framework that fuses information from heterogeneous sensors through representation learning and belief propagation, making target detection robust against malicious data injection attacks. Dr. Wang continues to collaborate with Dr. Qin's research group; their latest joint paper, "Bridging Symmetry and Robustness: On the Role of Equivariance in Enhancing Adversarial Robustness," appeared at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025).
Click here to read the NeurIPS 2025 paper.