Follow
Daniil Merkulov
Daniil Merkulov
Skoltech, MIPT
Verified email at phystech.edu - Homepage
Title
Cited by
Cited by
Year
Follow the bisector: a simple method for multi-objective optimization
A Katrutsa, D Merkulov, N Tursynbek, I Oseledets
arXiv preprint arXiv:2007.06937, 2020
62020
A new multi-objective approach to optimize irrigation using a crop simulation model and weather history
M Gasanov, D Merkulov, A Nikitin, S Matveev, N Stasenko, A Petrovskaia, ...
International Conference on Computational Science, 75-88, 2021
52021
Empirical Study of Extreme Overfitting Points of Neural Networks
D Merkulov, I Oseledets
Journal of Communications Technology and Electronics 64 (12), 1527–1534, 2020
52020
Memory-efficient backpropagation through large linear layers
D Bershatsky, A Mikhalev, A Katrutsa, J Gusak, D Merkulov, I Oseledets
arXiv preprint arXiv:2201.13195, 2022
42022
Stochastic gradient algorithms from ODE splitting perspective
D Merkulov, I Oseledets
arXiv preprint arXiv:2004.08981, 2020
32020
NAG-GS: Semi-Implicit, Accelerated and Robust Stochastic Optimizer
V Leplat, D Merkulov, A Katrutsa, D Bershatsky, O Tsymboi, I Oseledets
arXiv preprint arXiv:2209.14937, 2022
22022
Quantization of Large Language Models with an Overdetermined Basis
D Merkulov, D Cherniuk, A Rudikov, I Oseledets, E Muravleva, A Mikhalev, ...
arXiv preprint arXiv:2404.09737, 2024
2024
Another approach to build Lyapunov functions for the first order methods in the quadratic case
D Merkulov, I Oseledets
arXiv preprint arXiv:2310.15880, 2023
2023
Fast Line Search for Multi-Task Learning
A Filatov, D Merkulov
arXiv preprint arXiv:2110.00874, 2021
2021
Об ускоренных методах поиска канонического тензорного разложения
ДМ Меркулов, НК Тупица
Труды МФТИ 12 (4), 61-71, 2020
2020
The system can't perform the operation now. Try again later.
Articles 1–10