| Comparison of Accuracy and Scalability of Gauss--Newton and Alternating Least Squares for CANDECOMC/PARAFAC Decomposition N Singh, L Ma, H Yang, E Solomonik SIAM Journal on Scientific Computing 43 (4), C290-C311, 2021 | 27* | 2021 |
| Pruning Before Training May Improve Generalization, Provably H Yang, Y Liang, X Guo, L Wu, Z Wang arXiv preprint arXiv:2301.00335, 2023 | 14* | 2023 |
| Continuous Regular Functions AB Gorman, P Hieronymi, E Kaplan, R Meng, E Walsberg, Z Wang, ... Logical Methods in Computer Science 16, 2020 | 14 | 2020 |
| On the Neural Tangent Kernel Analysis of Randomly Pruned Neural Networks H Yang, Z Wang International Conference on Artificial Intelligence and Statistics, 1513-1553, 2023 | 10* | 2023 |
| Training dynamics of transformers to recognize word co-occurrence via gradient flow analysis H Yang, B Kailkhura, Z Wang, Y Liang Advances in Neural Information Processing Systems 37, 46047-46117, 2025 | 9 | 2025 |
| Neural Networks with Sparse Activation Induced by Large Bias: Tighter Analysis with Bias-Generalized NTK H Yang, Z Jiang, R Zhang, Y Liang, Z Wang Journal of Machine Learning Research 25 (321), 1-51, 2024 | 4* | 2024 |
| Transformers Provably Learn Two-Mixture of Linear Classification via Gradient Flow H Yang, Z Wang, JD Lee, Y Liang The Thirteenth International Conference on Learning Representations, 2025 | 3 | 2025 |
Zhangyang (Atlas) WangXTX Markets & University of Texas at AustinVerified email at utexas.edu
