Sledovať
Yichuan Deng
Yichuan Deng
Overená e-mailová adresa na: cs.washington.edu - Domovská stránka
Názov
Citované v
Citované v
Rok
Attention scheme inspired softmax regression
Y Deng, Z Li, Z Song
arXiv preprint arXiv:2304.10411, 2023
492023
Training overparametrized neural networks in sublinear time
Y Deng, H Hu, Z Song, O Weinstein, D Zhuo
arXiv preprint arXiv:2208.04508, 2022
282022
Discrepancy minimization in input-sparsity time
Y Deng, Z Song, O Weinstein
arXiv preprint arXiv:2210.12468, 2022
262022
Randomized and deterministic attention sparsification algorithms for over-parameterized feature dimension
Y Deng, S Mahadevan, Z Song
arXiv preprint arXiv:2304.04397, 2023
202023
Superiority of softmax: Unveiling the performance edge over linear attention
Y Deng, Z Song, T Zhou
arXiv preprint arXiv:2310.11685, 2023
162023
An improved sample complexity for rank-1 matrix sensing
Y Deng, Z Li, Z Song
arXiv preprint arXiv:2303.06895, 2023
162023
Unmasking transformers: A theoretical approach to data recovery via attention weights
Y Deng, Z Song, S Xie, C Yang
arXiv preprint arXiv:2310.12462, 2023
122023
Zero-th order algorithm for softmax attention optimization
Y Deng, Z Li, S Mahadevan, Z Song
2024 IEEE International Conference on Big Data (BigData), 24-33, 2024
102024
Attention is naturally sparse with gaussian distributed input
Y Deng, Z Song, C Yang
arXiv preprint arXiv:2404.02690, 2024
92024
Convergence of two-layer regression with nonlinear units
Y Deng, Z Song, S Xie
arXiv preprint arXiv:2308.08358, 2023
92023
Fast distance oracles for any symmetric norm
Y Deng, Z Song, O Weinstein, R Zhang
Advances in Neural Information Processing Systems 35, 7304-7317, 2022
82022
Solving tensor low cycle rank approximation
Y Deng, Y Gao, Z Song
2023 IEEE International Conference on Big Data (Big Data), 2023
72023
Dynamic kernel sparsifiers
Y Deng, W Jin, Z Song, X Sun, O Weinstein
arXiv preprint arXiv:2211.14825, 2022
72022
A nearly optimal size coreset algorithm with nearly linear time
Y Deng, Z Song, Y Wang, Y Yang
arXiv preprint arXiv:2210.08361, 2022
52022
Faster robust tensor power method for arbitrary order
Y Deng, Z Song, J Yin
arXiv preprint arXiv:2306.00406, 2023
42023
Enhancing stochastic gradient descent: A unified framework and novel acceleration methods for faster convergence
Y Deng, Z Song, C Yang
arXiv preprint arXiv:2402.01515, 2024
12024
Efficient Algorithm for Solving Hyperbolic Programs
Y Deng, Z Song, L Zhang, R Zhang
arXiv preprint arXiv:2306.07587, 2023
12023
Clustered Linear Contextual Bandits with Knapsacks
Y Deng, M Mamakos, Z Song
arXiv preprint arXiv:2308.10722, 2023
2023
Streaming kernel pca algorithm with small space
Y Deng, Z Song, Z Wang, H Zhang
arXiv preprint arXiv:2303.04555, 2023
2023
Systém momentálne nemôže vykonať operáciu. Skúste to neskôr.
Články 1–19