Urmăriți
Kevin Scaman
Kevin Scaman
Research scientist, INRIA
Adresă de e-mail confirmată pe inria.fr - Pagina de pornire
Titlu
Citat de
Citat de
Anul
Lipschitz regularity of deep neural networks: analysis and efficient estimation
A Virmaux, K Scaman
Advances in Neural Information Processing Systems 31, 2018
5342018
Optimal algorithms for smooth and strongly convex distributed optimization in networks
K Scaman, F Bach, S Bubeck, YT Lee, L Massoulié
Proceedings of the 34th International Conference on Machine Learning-Volume …, 2017
3542017
Optimal algorithms for non-smooth distributed optimization in networks
K Scaman, F Bach, S Bubeck, L Massoulié, YT Lee
Advances in Neural Information Processing Systems 31, 2018
1892018
Coloring graph neural networks for node disambiguation
G Dasoulas, LD Santos, K Scaman, A Virmaux
arXiv preprint arXiv:1912.06058, 2019
832019
Optimal convergence rates for convex distributed optimization in networks
K Scaman, F Bach, S Bubeck, YT Lee, L Massoulié
Journal of Machine Learning Research 20 (159), 1-31, 2019
832019
Multivariate Hawkes processes for large-scale inference
R Lemonnier, K Scaman, A Kalogeratos
Proceedings of the AAAI conference on artificial intelligence 31 (1), 2017
452017
Suppressing epidemics in networks using priority planning
K Scaman, A Kalogeratos, N Vayatis
IEEE Transactions on Network Science and Engineering 3 (4), 271-285, 2016
36*2016
Lipschitz normalization for self-attention layers with application to graph neural networks
G Dasoulas, K Scaman, A Virmaux
International Conference on Machine Learning, 2456-2466, 2021
312021
Robustness analysis of non-convex stochastic gradient descent using biased expectations
K Scaman, C Malherbe
Advances in Neural Information Processing Systems 33, 16377-16387, 2020
262020
Tight bounds for influence in diffusion networks and application to bond percolation and epidemiology
R Lemonnier, K Scaman, N Vayatis
Advances in Neural Information Processing Systems 27, 2014
262014
Tight high probability bounds for linear stochastic approximation with fixed stepsize
A Durmus, E Moulines, A Naumov, S Samsonov, K Scaman, HT Wai
Advances in Neural Information Processing Systems 34, 30063-30074, 2021
222021
A greedy approach for dynamic control of diffusion processes in networks
K Scaman, A Kalogeratos, N Vayatis
2015 IEEE 27th International Conference on Tools with Artificial …, 2015
192015
On sample optimality in personalized collaborative and federated learning
M Even, L Massoulié, K Scaman
Advances in Neural Information Processing Systems 35, 212-225, 2022
14*2022
Anytime influence bounds and the explosive behavior of continuous-time diffusion networks
K Scaman, R Lemonnier, N Vayatis
Advances in Neural Information Processing Systems 28, 2015
142015
Sequential informed federated unlearning: Efficient and provable client unlearning in federated optimization
Y Fraboni, M Van Waerebeke, K Scaman, R Vidal, L Kameni, M Lorenzi
arXiv preprint arXiv:2211.11656, 2022
132022
Convergence rates of non-convex stochastic gradient descent under a generic lojasiewicz condition and local smoothness
K Scaman, C Malherbe, L Dos Santos
International conference on machine learning, 19310-19327, 2022
132022
Ego-based entropy measures for structural representations on graphs
G Dasoulas, G Nikolentzos, K Scaman, A Virmaux, M Vazirgiannis
ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and …, 2021
8*2021
Theoretical limits of pipeline parallel optimization and application to distributed deep learning
I Colin, L Dos Santos, K Scaman
Advances in Neural Information Processing Systems 32, 2019
82019
Improving hierarchical adversarial robustness of deep neural networks
A Ma, A Virmaux, K Scaman, J Lu
arXiv preprint arXiv:2102.09012, 2021
62021
Information diffusion and rumor spreading
A Kalogeratos, K Scaman, L Corinzia, N Vayatis
Cooperative and Graph Signal Processing, 651-678, 2018
62018
Sistemul nu poate realiza operația în acest moment. Încercați din nou mai târziu.
Articole 1–20