Presentation, SUSTech, Prof. Chao WANG's Seminar, Shenzhen, China
The rank based on the grid tensor matrix is used as the lower bound of the separation rank, and the theory of neural network is explained from the perspective of rank, including depth, width, and bottleneck. link
Presentation, SUSTech, Prof. Chao WANG's Seminar, Shenzhen, China
Introduced qMRI and the applications of three kinds of physics-driven learning methods. link
Presentation, SUSTech, Prof. Chao WANG's Seminar, Shenzhen, China
Introduced three modes of EM image processing, and focused on the unrolling-based method of physics-embedded model.link
Presentation, SUSTech, Prof. Chao WANG's Seminar, Shenzhen, China
Introduced the classic generative model foundations VAEs, GANs and Diffusion Model, and then used these theories to explain physics-driven applications of generative models. link
Presentation, SUSTech, Prof. Chao WANG's Seminar, Shenzhen, China
Introduced the concept of equivariant and confirmed the role of equivariant in deep learning network construction and learning. link