Следене
Fangshuo Liao
Fangshuo Liao
Ph.D. Student, Rice University
Потвърден имейл адрес: rice.edu - Начална страница
Заглавие
Позовавания
Позовавания
Година
Scissorhands: Exploiting the persistence of importance hypothesis for llm kv cache compression at test time
Z Liu, A Desai, F Liao, W Wang, V Xie, Z Xu, A Kyrillidis, A Shrivastava
Advances in Neural Information Processing Systems 36, 2024
372024
On the convergence of shallow neural network training with randomly masked neurons
F Liao, A Kyrillidis
arXiv preprint arXiv:2112.02668, 2021
182021
GIST: Distributed training for large-scale graph convolutional networks
CR Wolfe, J Yang, F Liao, A Chowdhury, C Dun, A Bayer, S Segarra, ...
Journal of Applied and Computational Topology, 1-53, 2023
122023
LOFT: Finding lottery tickets through filter-wise training
Q Wang, C Dun, F Liao, C Jermaine, A Kyrillidis
International Conference on Artificial Intelligence and Statistics, 6498-6526, 2023
32023
How much pre-training is enough to discover a good subnetwork?
CR Wolfe, Q Wang, JL Kim, A Kyrillidis
22021
Strong Lottery Ticket Hypothesis with –perturbation
Z Xiong, F Liao, A Kyrillidis
International Conference on Artificial Intelligence and Statistics, 6879-6902, 2023
12023
Provable Accelerated Convergence of Nesterov’s Momentum for Deep ReLU Neural Networks
F Liao, A Kyrillidis
International Conference on Algorithmic Learning Theory, 732-784, 2024
2024
On the Error-Propagation of Inexact Deflation for Principal Component Analysis
F Liao, JL Kim, C Barnum, A Kyrillidis
arXiv preprint arXiv:2310.04283, 2023
2023
Accelerated Convergence of Nesterov's Momentum for Deep Neural Networks under Partial Strong Convexity
F Liao, A Kyrillidis
arXiv preprint arXiv:2306.08109, 2023
2023
Strong Lottery Ticket Hypothesis with –perturbation
F Liao, Z Xiong, A Kyrillidis
OPT 2022: Optimization for Machine Learning (NeurIPS 2022 Workshop), 2022
2022
Системата не може да изпълни операцията сега. Опитайте отново по-късно.
Статии 1–10