Dmitry Vetrov
Dmitry Vetrov
Professor of Computer Science at Constructor University, Bremen
Потвърден имейл адрес: constructor.university - Начална страница
Averaging weights leads to wider optima and better generalization
P Izmailov, D Podoprikhin, T Garipov, D Vetrov, AG Wilson
arXiv preprint arXiv:1803.05407, 2018
Variational dropout sparsifies deep neural networks
D Molchanov, A Ashukha, D Vetrov
International conference on machine learning, 2498-2507, 2017
Tensorizing neural networks
A Novikov, D Podoprikhin, A Osokin, DP Vetrov
Advances in neural information processing systems 28, 2015
A simple baseline for bayesian uncertainty in deep learning
WJ Maddox, P Izmailov, T Garipov, DP Vetrov, AG Wilson
Advances in neural information processing systems 32, 2019
Loss surfaces, mode connectivity, and fast ensembling of dnns
T Garipov, P Izmailov, D Podoprikhin, DP Vetrov, AG Wilson
Advances in neural information processing systems 31, 2018
Evaluation of stability of k-means cluster ensembles with respect to random initialization
LI Kuncheva, DP Vetrov
IEEE transactions on pattern analysis and machine intelligence 28 (11), 1798 …, 2006
Spatially Adaptive Computation Time for Residual Networks
M Figurnov, M Collins, Y Zhu, L Zhang, J Huang, DP Vetrov, ...
Pitfalls of in-domain uncertainty estimation and ensembling in deep learning
A Ashukha, A Lyzhov, D Molchanov, D Vetrov
arXiv preprint arXiv:2002.06470, 2020
Entangled conditional adversarial autoencoder for de novo drug discovery
D Polykovskiy, A Zhebrak, D Vetrov, Y Ivanenkov, V Aladinskiy, ...
Molecular pharmaceutics 15 (10), 4398-4405, 2018
Structured bayesian pruning via log-normal multiplicative noise
K Neklyudov, D Molchanov, A Ashukha, DP Vetrov
Advances in Neural Information Processing Systems 30, 2017
Ultimate tensorization: compressing convolutional and fc layers alike
T Garipov, D Podoprikhin, A Novikov, D Vetrov
arXiv preprint arXiv:1611.03214, 2016
Breaking sticks and ambiguities with adaptive skip-gram
S Bartunov, D Kondrashkin, A Osokin, D Vetrov
artificial intelligence and statistics, 130-138, 2016
Perforatedcnns: Acceleration through elimination of redundant convolutions
M Figurnov, A Ibraimova, DP Vetrov, P Kohli
Advances in neural information processing systems 29, 2016
Subspace inference for Bayesian deep learning
P Izmailov, WJ Maddox, P Kirichenko, T Garipov, D Vetrov, AG Wilson
Uncertainty in Artificial Intelligence, 1169-1179, 2020
Controlling overestimation bias with truncated mixture of continuous distributional quantile critics
A Kuznetsov, P Shvechikov, A Grishin, D Vetrov
International Conference on Machine Learning, 5556-5566, 2020
Variational autoencoder with arbitrary conditioning
O Ivanov, M Figurnov, D Vetrov
arXiv preprint arXiv:1806.02382, 2018
Fast adaptation in generative models with generative matching networks
S Bartunov, DP Vetrov
arXiv preprint arXiv:1612.02192, 2016
Conditional generators of words definitions
A Gadetsky, I Yakubovskiy, D Vetrov
arXiv preprint arXiv:1806.10090, 2018
Predictive model for bottomhole pressure based on machine learning
P Spesivtsev, K Sinkov, I Sofronov, A Zimina, A Umnov, R Yarullin, ...
Journal of Petroleum Science and Engineering 166, 825-841, 2018
Greedy policy search: A simple baseline for learnable test-time augmentation
A Lyzhov, Y Molchanova, A Ashukha, D Molchanov, D Vetrov
Conference on uncertainty in artificial intelligence, 1308-1317, 2020
Системата не може да изпълни операцията сега. Опитайте отново по-късно.
Статии 1–20