Skip to main content

Qiang Ye

Research Interests:
applied and computational mathematics
Education

Ph.D., University of Calgary, 1989
M.Sc, Chinese Academy of Sciences, 1986
B.Sc., University of Science and Technology of China, 1983

Research

My recent research results are in the areas of deep learning and numerical analysis.

 

 

Selected Publications:

SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Tasks, (with V. Zadorozhnyy and K. Koishida), INTERSPEECH 2023.

Novel Molecular Representations using Neumann-Cayley Orthogonal Gated Recurrent Unit, (with E. Mucllari, V. Zadorozhnyy, and D. Nguyen) , Journal of Chemical Information and Modeling, 2023, 63, 9, 2656–2666.

Batch Normalization Preconditioning for Neural Network Training, (with Susanna Lange and Kyle Helfrich),  Journal of Machine Learning Research, 23(72):1-41, 2022.

Symmetry-Structured Convolutional Neural Networks, (with K.D. Gayan Maduranga and Vasily Zadorozhnyy),  Neural Computing and Applications.35, 4421–4434 (2023).

Stochastic Gradient Descent with Nonlinear Conjugate Gradient-Style Adaptive Momentum, (with Bao Wang), IEEE Transactions on Neural Networks and Learning Systems. 2023.

Adaptive Weighted Discriminator for Training Generative Adversarial Networks, (with Vasily Zadorozhnyy and Qiang Cheng), Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 4781-4790.

Eigenvalue Normalized Recurrent Neural Networks for Short Term Memory, (with K. Helfrich), (arXiv:1911.07964), Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence, Vol 34 No 04: Pages 4115-4122, 2020.

Orthogonal Recurrent Neural Networks with Scaled Cayley Transform, (with K. Helfrich and D. Willmott), Proceedings of the 35th International Conference on Machine Learning, ICML 2018, PMLR 80:1969-1978, 2018.

Preconditioning for Accurate Solutions of Ill-Conditioned Linear Systems, Numerical Linear Algebra with Applications, 27(2020):e2315.

Error Bounds for the Krylov Subspace methods for Computations of Matrix Exponentials, (with H. Wang), SIAM J. Matrix Anal. Appl., 38(1), 155–187.

Accurate Inverses for Computing Eigenvalues of Extremely Ill-Conditioned Matrices and Differential Operators, Math. Comp. 87 (2018), 237-259