?
Rates of convergence for density estimation with generative adversarial networks
Journal of Machine Learning Research. 2024. Vol. 25. No. 29. P. 1-47.
In this work we undertake a thorough study of the non-asymptotic properties of the vanilla generative adversarial networks (GANs). We prove an oracle inequality for the Jensen-Shannon (JS) divergence between the underlying density 𝗉∗ and the GAN estimate with a significantly better statistical error term compared to the previously known results. The advantage of our bound becomes clear in application to nonparametric density estimation. We show that the JS-divergence between the GAN estimate and 𝗉∗ decays as fast as (logn/n)2β/(2β+d), where n is the sample size and β determines the smoothness of 𝗉∗. This rate of convergence coincides (up to logarithmic factors) with minimax optimal for the considered class of densities.
Publication based on the results of:
Kadurin A., Nikolenko S. I., Khrabrov K. et al., Molecular Pharmaceutics 2017 Vol. 14 No. 9 P. 3098-3104
Deep generative adversarial networks (GANs) are the emerging technology in drug discovery and biomarker development. In our recent work, we demonstrated a proof-of-concept of implementing deep generative adversarial autoencoder (AAE) to identify new molecular fingerprints with predefined anticancer properties. Another popular generative model is the variational autoencoder (VAE), which is based on deep neural architectures. ...
Added: March 14, 2018
Егоров Е. А., Rogachev A., Доклады Российской академии наук. Математика, информатика, процессы управления (ранее - Доклады Академии Наук. Математика) 2023 Т. 514 № 2 С. 49-59
When using Wasserstein GAN loss function for training generative adversarial networks (GAN), it is theoretically necessary to limit the discriminators’ expressive power (so called discriminator normalization). Such limitation increases the stability of GAN training at the expense of a less expressive final model. Spectral normalization is one of the normalization algorithms that involves applying a ...
Added: November 30, 2023
Sergeev F., Jain N., Kostenkov G. et al., Journal of Physics: Conference Series 2021 No. 1740 Article 012028
Modern experiments in high-energy physics require an increasing amount of simulated data. Monte-Carlo simulation of calorimeter responses is by far the most computationally expensive part of such simulations. Recent works have shown that the application of generative neural networks to this task can significantly speed up the simulations while maintaining an appropriate degree of accuracy. ...
Added: January 25, 2021
Polykovskiy D., Zhebrak A., Sanchez-Lengeling B. et al., Frontiers in Pharmacology 2020 Vol. 11 P. 1-10
Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervized predictive models in the downstream tasks. While there are plenty of generative models, it is ...
Added: April 21, 2021
Aaij R., Abdelmotteleb A. S., Abellan Beteta C. et al., The European Physical Journal C - Particles and Fields 2023 Vol. 83 Article 543
https://link.springer.com/article/10.1140/epjc/s10052-023-11673-x#Abs1 ...
Added: December 4, 2023
Borchmann D., Hanika T., Obiedkov S., Discrete Applied Mathematics 2020 Vol. 273 P. 30-42
We propose an algorithm for learning the Horn envelope of an arbitrary domain using an expert, or an oracle, capable of answering certain types of queries about this domain. Attribute exploration from formal concept analysis is a procedure that solves this problem, but the number of queries it may ask is exponential in the size ...
Added: October 29, 2019
Springer, 2012
Added: January 29, 2013
М. : Издательский центр «Российский государственный гуманитарный университет», 2019
Сборник включает 27 докладов международной конференции по компьютерной лингвистике и интеллектуальным технологиям «Диалог 2019», не вошедшие в ежегодник «Компьютерная лингвистика и интеллектуальные технологии», но рекомендованные Программным Комитетом к представлению на конференции. Для специалистов в области теоретической и прикладной лингвистики и интеллектуальных технологий. ...
Added: December 10, 2019
Toldova S., Azerkovich I., Гришина Ю. et al., / НИУ ВШЭ. Series WP BRP "Linguistics". 2015.
Building benchmark corpora in the domain of coreference and anaphora resolution is an important task for developing and evaluating NLP systems and models. Our study is aimed at assessing the feasibility of enhancing corpora with information about coreference relations. The annotation procedure includes identification of text segments that are subjects to annotation (markables), marking their ...
Added: December 15, 2015
Kiselyova N. N., Dudarev V.A., Korzhuev M. A., Inorganic Materials: Applied Research 2016 Vol. 7 No. 1 P. 34-39
A database (DB) on the bandgap of inorganic substances available via the Internet (http://bg.imetdb.ru) was developed for the information service of specialists in the sphere of inorganic chemistry and materials science. The DB is integrated with other information systems on the properties of inorganic substances and materials, which provides the search of a wide range ...
Added: February 23, 2016
Litvin Y. V., Абрамов И. В., Технологии техносферной безопасности 2016 № 66
Advanced approach to the assessment of a random time of arrival fire fighting calculation on the object of protection, the time of their employment and the free combustion. There is some quantitative assessments with the review of analytical methods and simulation ...
Added: August 27, 2016
Gostev I. M., М. : Юрайт, 2016
В настоящее время компьютерные науки стремительно развиваются. Новые версии операционных систем появляются каждые полтора-два года, поэтому было принято решение о включении в данную книгу такого материала, который не будет устаревать. Содержание учебника представляет собой некоторые наиболее общие принципы построения операционных систем, которые были разработаны более 50 лет назад и практически не изменились за прошедшее время. ...
Added: October 13, 2009
Beklemishev L. D., Оноприенко А. А., Математический сборник 2015 Т. 206 № 9 С. 3-20
We formulate some term rewriting systems in which the number of computation steps is finite for each output, but this number cannot be bounded by a provably total computable function in Peano arithmetic PA. Thus, the termination of such systems is unprovable in PA. These systems are derived from an independent combinatorial result known as the Worm ...
Added: March 13, 2016
Ружицкая Д. Д., САМОЙЛЕНКО А. А., Иванов А. Д. et al., Optoelectronics, Instrumentation and Data Processing 2017 Vol. 54 No. 1 P. 1-8
This paper presents an algorithm for processing the transmission spectra of whisperinggallery optical microcavities for use as a nanoparticle detector. The algorithm is based on the broadening of the microcavity resonance curve during precipitation of nanoparticles on the microcavity surface. Experimental results on the detection of particles are compared with Langmuir adsorption theory. The contribution ...
Added: May 25, 2018
Dneprovskaya N., Шевцова И. В., Бизнес-информатика 2023 Т. 17 № 2 С. 20-40
The purpose of this study is a conceptual description of the implementation of knowledge management systems (KMS) as a mechanism for universities’ strategic development. Knowledge management (KM) practice from around the world proved the positive influence of KMS on productivity of educational institutions. The theoretical provisions and concept for KMS are determined based on an ...
Added: August 2, 2023
Каз. : Издательство «Фэн» Академии наук Республики Татарстан, 2013
Материалы и доклады Шестой Всероссийской научно-практической конференции по имитацонному моделированию и его применению в науке и промышленности. ...
Added: December 14, 2013
Singapore : World Scientific, 1999
Added: January 26, 2014
Shuranov E., / Cornell University. Series Computer Science "arxiv.org". 2021.
Text encodings from automatic speech recognition (ASR) transcripts and audio representations have shown promise in speech emotion recognition (SER) ever since. Yet, it is challenging to explain the effect of each information stream on the SER systems. Further, more clarification is required for analysing the impact of ASR's word error rate (WER) on linguistic emotion ...
Added: February 14, 2023
Aaij R., Abdelmotteleb A. S., Abellán Beteta C. et al., Physical Review Letters 2022 Vol. 129 No. 9 Article 091801
The first study of the angular distribution of μþμ− pairs produced in the forward rapidity region via the Drell-Yan reaction pp → γ=Z þ X → lþl− þ X is presented, using data collected with the LHCb detector at a center-of-mass energy of 13 TeV, corresponding to an integrated luminosity of 5.1 fb−1. The coefficients ...
Added: December 29, 2022
Svetlana Maltseva, Kornilov V., Barakhnin V. et al., Complexity 2022 Vol. 2022 Article 5714395
We can observe self-organization properties in various systems. However, modern networked dynamical sociotechnical systems have some features that allow for realizing the benefits of self-organization in a wide range of systems in economic and social areas. The review examines the general principles of self-organized systems, as well as the features of the implementation of self-organization ...
Added: April 18, 2022
Karpov V. E., Karpova I. P., Procedia Engineering 2015 Vol. 100 P. 1459-1468
Work solutions are proposed for problems of leader definition and role distribution in homogeneous groups of robots. It is shown that transition from a swarm to a collective of robots with hierarchical organization is possible using exclusively local interaction. The local revoting algorithm is central to the procedure for choice of leader while redistribution of roles can ...
Added: March 14, 2015
М. : National Instruments Russia, 2017
Содержание сборника составляют доклады с результатами оригинальных исследований и технических решений, ранее не публиковавшиеся. Мы надеемся, что предлагаемый сборник окажется полезным для специалистов, работающих в различных областях науки и техники, для широкого круга преподавателей, аспирантов и студентов ВУЗов, а также для преподавателей средних школ и технических колледжей. ...
Added: May 10, 2017
Furmanov K. K., Nikol'skii I. M., Computational Mathematics and Modeling 2016 Vol. 27 No. 2 P. 247-253
Added: December 22, 2016
Morozov I., Norman G. E., Insepov Z. et al., Physical Review Special Topics - Accelerators and Beams 2012 Vol. 15 P. 053501
This paper describes the surface environment of the dense plasma arcs that damage rf accelerators, tokamaks, and other high gradient structures. We simulate the dense, nonideal plasma sheath near a metallic surface using molecular dynamics (MD) to evaluate sheaths in the non-Debye region for high density, low temperature plasmas. We use direct two-component MD simulations ...
Added: October 28, 2013