Follow
Soufiane Hayou
Soufiane Hayou
Simons Institute for the Theory of Computing, UC Berkeley
Verified email at berkeley.edu - Homepage
Title
Cited by
Cited by
Year
On the impact of the activation function on deep neural networks training
S Hayou, A Doucet, J Rousseau
36th International Conference on Machine Learning (ICML 2019), 2019
2332019
On the selection of initialization and activation function for deep neural networks
S Hayou, A Doucet, J Rousseau
arXiv preprint arXiv:1805.08266, 2018
852018
Stable ResNet
S Hayou, E Clerico, B He, G Deligiannidis, A Doucet, J Rousseau
24th International Conference on Artificial Intelligence and Statistics …, 2021
462021
Robust Pruning at Initialization
S Hayou, JF Ton, A Doucet, YW Teh
International Conference on Learning Representations (ICLR 2021), 2021
422021
Mean-field Behaviour of Neural Tangent Kernel for Deep Neural Networks
S Hayou, A Doucet, J Rousseau
arXiv preprint arXiv:1905.13654, 2019
282019
On the infinite-depth limit of finite-width neural networks
S Hayou
Transactions on Machine Learning Research (arXiv:2210.00688), 2022
172022
Feature learning and signal propagation in deep neural networks
Y Lou, CE Mingard, S Hayou
International Conference on Machine Learning, 14248-14282, 2022
122022
Pruning untrained neural networks: Principles and analysis
S Hayou, JF Ton, A Doucet, Y Whye Teh
arXiv e-prints, arXiv: 2002.08797, 2020
122020
Training dynamics of deep networks using stochastic gradient descent via neural tangent kernel
S Hayou, A Doucet, J Rousseau
arXiv preprint arXiv:1905.13654, 2019
122019
On the impact of the activation function on deep neural networks training 2019
S Hayou, A Doucet, J Rousseau
arXiv preprint arXiv:1902.06853, 1902
11*1902
Regularization in ResNet with Stochastic Depth
S Hayou, F Ayed
NeurIPS 2021, arXiv:2106.03091, 2021
102021
Width and Depth Limits Commute in Residual Networks
S Hayou, G Yang
ICML 2023 (arXiv preprint arXiv:2302.00453), 2023
82023
The curse of depth in kernel regime
S Hayou, A Doucet, J Rousseau
I (Still) Can't Believe It's Not Better! Workshop at NeurIPS 2021, 41-47, 2022
62022
Data pruning and neural scaling laws: fundamental limitations of score-based algorithms
F Ayed, S Hayou
Transactions on Machine Learning Research 2023. (arXiv preprint arXiv:2302 …, 2023
52023
LoRA+: Efficient Low Rank Adaptation of Large Models
S Hayou, N Ghosh, B Yu
arXiv preprint arXiv:2402.12354, 2024
42024
Tensor Programs VI: Feature Learning in Infinite-Depth Neural Networks
G Yang, D Yu, C Zhu, S Hayou
ICLR 2024, 2023
42023
Connecting Optimization and Generalization via Gradient Flow Path Length
F Liu, H Yang, S Hayou, Q Li
arXiv preprint arXiv:2202.10670, 2022
42022
Feature Learning in Infinite Depth Neural Networks
G Yang, D Yu, C Zhu, S Hayou
The Twelfth International Conference on Learning Representations, 2023
32023
Leave-one-out distinguishability in machine learning
J Ye, A Borovykh, S Hayou, R Shokri
ICLR 2024, arXiv preprint arXiv:2309.17310, 2023
22023
From Optimization Dynamics to Generalization Bounds via Łojasiewicz Gradient Inequality
F Liu, H Yang, S Hayou, Q Li
Transactions on Machine Learning Research, 2022
22022
The system can't perform the operation now. Try again later.
Articles 1–20