Παρακολούθηση
Rustem Islamov
Rustem Islamov
PhD student, University of Basel
Η διεύθυνση ηλεκτρονικού ταχυδρομείου έχει επαληθευτεί στον τομέα unibas.ch - Αρχική σελίδα
Τίτλος
Παρατίθεται από
Παρατίθεται από
Έτος
FedNL: Making Newton-type methods applicable to federated learning
M Safaryan, R Islamov, X Qian, P Richtárik
International Conference on Machine Learning (ICML 2022), 2021
922021
Distributed second order methods with fast rates and compressed communication
R Islamov, X Qian, P Richtárik
International conference on machine learning, 4617-4628, 2021
622021
Basis matters: better communication-efficient second order methods for federated learning
X Qian, R Islamov, M Safaryan, P Richtárik
24th International Conference on Artificial Intelligence and Statistics …, 2021
252021
AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms
R Islamov, M Safaryan, D Alistarh
the 27th International Conference on Artificial Intelligence and Statistics, 2023
132023
Distributed Newton-type methods with communication compression and bernoulli aggregation
R Islamov, X Qian, S Hanzely, M Safaryan, P Richtárik
Transactions on Machine Learning Research, 2022
122022
Adaptive compression for communication-efficient distributed training
M Makarenko, E Gasanov, R Islamov, A Sadiev, P Richtárik
Transactions on Machine Learning Research, 2022
102022
Partially personalized federated learning: Breaking the curse of data heterogeneity
K Mishchenko, R Islamov, E Gorbunov, S Horváth
arXiv preprint arXiv:2305.18285, 2023
92023
Clip21: Error feedback for gradient clipping
S Khirirat, E Gorbunov, S Horváth, R Islamov, F Karray, P Richtárik
arXiv preprint arXiv:2305.18929, 2023
52023
EControl: Fast Distributed Optimization with Compression and Error Control
Y Gao, R Islamov, S Stich
International Conference on Learning Representations, 2023
42023
Towards Faster Decentralized Stochastic Optimization with Communication Compression
R Islamov, Y Gao, S Stich
ICLR 2025 - International Conference on Learning Representations, 2025
1*2025
Adaptive Methods through the Lens of SDEs: Theoretical Insights on the Role of Noise
EM Compagnoni, T Liu, R Islamov, FN Proske, A Orvieto, A Lucchi
arXiv preprint arXiv:2411.15958, 2024
12024
Loss Landscape Characterization of Neural Networks without Over-Parametrization
R Islamov, N Ajroldi, A Orvieto, A Lucchi
arXiv preprint arXiv:2410.12455, 2024
12024
Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs
EM Compagnoni, R Islamov, FN Proske, A Lucchi
The 28th International Conference on Artificial Intelligence and Statistics, 0
Δεν είναι δυνατή η εκτέλεση της ενέργειας από το σύστημα αυτή τη στιγμή. Προσπαθήστε ξανά αργότερα.
Άρθρα 1–13