Zobrazeno 1 - 10
of 21
pro vyhledávání: '"Anuran Makur"'
Publikováno v:
IEEE Open Journal of Control Systems, Vol 2, Pp 380-395 (2023)
Several recent empirical studies demonstrate that important machine learning tasks such as training deep neural networks, exhibit a low-rank structure, where most of the variation in the loss function occurs only in a few directions of the input spac
Externí odkaz:
https://doaj.org/article/6ea85ff32aa7488f883684b55870207f
Autor:
Anuran Makur, Lizhong Zheng
Publikováno v:
Problems of Information Transmission. 56:103-156
Contraction coefficients are distribution dependent constants that are used to sharpen standard data processing inequalities for f-divergences (or relative f-entropies) and produce so-called “strong” data processing inequalities. For any bivariat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d44e03c4802ea980babb9633cad2b0c7
https://doi.org/10.1134/s0032946020020015
https://doi.org/10.1134/s0032946020020015
Publikováno v:
ISIT
We investigate the problem of broadcasting a bit on a 2D regular grid. Consider a directed acyclic graph with the structure of a 2D regular grid, which has a single source vertex $X$ at layer 0, and $k+1$ vertices at distance of $k\geq 1$ from $X$ at
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::12573ec1a3c9eb6a4375ea0d28ead685
https://doi.org/10.1109/isit45174.2021.9518174
https://doi.org/10.1109/isit45174.2021.9518174
We introduce a new opinion dynamics model where a group of agents holds two kinds of opinions: inherent and declared. Each agent's inherent opinion is fixed and unobservable by the other agents. At each time step, agents broadcast their declared opin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6a2709145e84cee5300ded00dd4d0a6b
http://arxiv.org/abs/2104.11172
http://arxiv.org/abs/2104.11172
In many contemporary and emerging applications of machine learning and statistical inference, the phenomena of interest are characterized by variables defined over large alphabets. This increasing size of both the data and the number of inferences, a
Publikováno v:
ISIT
A modal decomposition is a useful tool that deconstructs the statistical dependence between two random variables by decomposing their joint distribution into orthogonal modes. Historically, modal decompositions have played important roles in statisti
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::644f272b5a3e10c18bc82691ce468706
https://doi.org/10.1109/isit44484.2020.9174093
https://doi.org/10.1109/isit44484.2020.9174093
Autor:
Anuran Makur
Publikováno v:
ISIT
The "permutation channel" model is a convenient representation of certain communication networks, where packets are not indexed and delivered out-of-order, and closely resembles models of DNA based storage systems. It consists of a standard discrete
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ea2441e7f8d0d633a7bdd1b942f6f907
https://doi.org/10.1109/isit44484.2020.9174455
https://doi.org/10.1109/isit44484.2020.9174455
We study a specialization of the problem of broadcasting on directed acyclic graphs, namely, broadcasting on 2D regular grids. Consider a 2D regular grid with source vertex $X$ at layer $0$ and $k+1$ vertices at layer $k\geq 1$, which are at distance
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b7cf62e92c91d758035dec5b0d58cb98
Publikováno v:
ISIT
MIT web domain
MIT web domain
© 2019 IEEE. We study a generalization of the problem of broadcasting on trees to the setting of directed acyclic graphs (DAGs). At time 0, a source vertex X transmits a uniform bit along binary symmetric channels (BSCs) to a set of vertices called
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::79b3ce0fe20a016eb491f72341c21b29
https://doi.org/10.1109/isit.2019.8849393
https://doi.org/10.1109/isit.2019.8849393
Autor:
Anuran Makur
Publikováno v:
Allerton
In this paper, we describe and study the permutation channel model, which constitutes a discrete memoryless channel (DMC) followed by a random permutation block that reorders the output codeword of the DMC. This model naturally emerges in the context
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::86059025f2b6a2292abdab22268fc5cd
https://doi.org/10.1109/allerton.2018.8636070
https://doi.org/10.1109/allerton.2018.8636070