A review of neural architecture search D Baymurzina, E Golikov, M Burtsev
Neurocomputing 474, 82-93, 2022
69 2022 Two-jet astrosphere model: effect of azimuthal magnetic field EA Golikov, VV Izmodenov, DB Alexashov, NA Belov
Monthly Notices of the Royal Astronomical Society 464 (1), 1065-1076, 2017
16 2017 Differentiable lower bound for expected BLEU score V Zhukov, E Golikov, M Kretov
arXiv preprint arXiv:1712.04708, 2017
15 2017 Towards a general theory of infinite-width limits of neural classifiers E Golikov
International Conference on Machine Learning, 3617-3626, 2020
10 2020 Feature Learning in -regularized DNNs: Attraction/Repulsion and Sparsity A Jacot, E Golikov, C Hongler, F Gabriel
Advances in Neural Information Processing Systems 35, 6763-6774, 2022
9 2022 Neural tangent kernel: A survey E Golikov, E Pokonechnyy, V Korviakov
arXiv preprint arXiv:2208.13614, 2022
9 2022 Non-gaussian tensor programs E Golikov, G Yang
Advances in Neural Information Processing Systems 35, 21521-21533, 2022
8 2022 Dynamically stable infinite-width limits of neural classifiers EA Golikov
arXiv preprint arXiv:2006.06574, 2020
8 2020 Two-jet structure of the flow produced by magnetized hypersonic spherical source into the steady unmagnetized medium EA Golikov, VV Izmodenov, DB Alexashov
Journal of Physics: Conference Series 815 (1), 012035, 2017
8 2017 Notes on Deep Learning Theory EA Golikov
arXiv preprint arXiv:2012.05760, 2020
1 2020 Quadratic number of nodes is sufficient to learn a dataset via gradient descent B Das, E Golikov
arXiv preprint arXiv:1911.05402, 2019
1 2019 An Essay on Optimization Mystery of Deep Learning E Golikov
arXiv preprint arXiv:1905.07187, 2019
2019 Embedding-reparameterization procedure for manifold-valued latent variables in generative models E Golikov, M Kretov
arXiv preprint arXiv:1812.02769, 2018
2018 Using stochastic computation graphs formalism for optimization of sequence-to-sequence model E Golikov, V Zhukov, M Kretov
arXiv preprint arXiv:1711.07724, 2017
2017 Infinitely Wide Nets E Golikov
Towards a General Theory of Infinite-Width Limits of Neural Classifiers: Supplementary Material EA Golikov