Maksim Velikanov
Maksim Velikanov
Researcher, Technology Innovation Institute
Bestätigte E-Mail-Adresse bei
Zitiert von
Zitiert von
Three-particle electron-hole complexes in two-dimensional electron systems
VA Kuznetsov, LV Kulik, MD Velikanov, AS Zhuravlev, AV Gorbunov, ...
Physical Review B 98 (20), 205303, 2018
Explicit loss asymptotics in the gradient descent training of neural networks
M Velikanov, D Yarotsky
Advances in Neural Information Processing Systems 34, 2570-2582, 2021
A view of mini-batch SGD via generating functions: conditions of convergence, phase transitions, benefit from negative momenta
M Velikanov, D Kuznedelev, D Yarotsky
arXiv preprint arXiv:2206.11124, 2022
Tight convergence rate bounds for optimization under power law spectral conditions
M Velikanov, D Yarotsky
Journal of Machine Learning Research 25 (81), 1-78, 2024
Embedded Ensembles: infinite width limit and operating regimes
M Velikanov, RV Kail, I Anokhin, R Vashurin, M Panov, A Zaytsev, ...
International Conference on Artificial Intelligence and Statistics, 3138-3163, 2022
Efficient Conformal Prediction under Data Heterogeneity
V Plassier, N Kotelevskii, A Rubashevskii, F Noskov, M Velikanov, ...
International Conference on Artificial Intelligence and Statistics, 4879-4887, 2024
Generalization error of spectral algorithms
M Velikanov, M Panov, D Yarotsky
arXiv preprint arXiv:2403.11696, 2024
Proton fraction in neutron star matter: dynamical mean-field approach
M Velikanov, AN Rubtsov, B Krippa
New Journal of Physics 23 (3), 033015, 2021
Tight Convergence Rate Bounds for First Order Optimization Under Power Law Spectral Conditions
M Velikanov, D Yarotsky
Insights into neutron star cores from the theory of correlated quantum materials
M Velikanov
Das System kann den Vorgang jetzt nicht ausführen. Versuchen Sie es später erneut.
Artikel 1–10