Zobrazeno 1 - 10
of 938
pro vyhledávání: '"Barron, Andrew R."'
Autor:
Takeuchi, Jun'ichi, Barron, Andrew R.
We study the problems of data compression, gambling and prediction of a sequence $x^n=x_1x_2...x_n$ from an alphabet ${\cal X}$, in terms of regret and expected regret (redundancy) with respect to various smooth families of probability distributions.
Externí odkaz:
http://arxiv.org/abs/2406.17929
Minimum Description Length (MDL) estimators, using two-part codes for universal coding, are analyzed. For general parametric families under certain regularity conditions, we introduce a two-part code whose regret is close to the minimax regret, where
Externí odkaz:
http://arxiv.org/abs/2311.03852
Autor:
Koide, Yoshihiro, Barron, Andrew R.
Publikováno v:
Main Group Metal Chemistry, Vol 18, Iss 8, Pp 405-414 (1995)
Externí odkaz:
https://doaj.org/article/5d9255d4df8c4b6fabcd68cde80f6f7b
Publikováno v:
In International Journal of Hydrogen Energy 22 March 2024 60:28-45
Autor:
Khodabakhshi, Saeed, Fulvio, Pasquale F., Walton, Krista S., Kiani, Sajad, Niu, Yubiao, Palmer, Richard E., Barron, Andrew R., Andreoli, Enrico
Publikováno v:
In Carbon March 2024 221
Publikováno v:
In Materials Science in Semiconductor Processing 1 August 2023 162
Autor:
Barron, Andrew R., Klusowski, Jason M.
For any ReLU network there is a representation in which the sum of the absolute values of the weights into each node is exactly $1$, and the input layer variables are multiplied by a value $V$ coinciding with the total variation of the path weights.
Externí odkaz:
http://arxiv.org/abs/1902.00800
Autor:
Barron, Andrew R., Klusowski, Jason M.
It has been experimentally observed in recent years that multi-layer artificial neural networks have a surprising ability to generalize, even when trained with far more parameters than observations. Is there a theoretical basis for this? The best ava
Externí odkaz:
http://arxiv.org/abs/1809.03090
Autor:
Cheraghian Goshtasp, Wistuba Michael P., Kiani Sajad, Behnood Ali, Afrand Masoud, Barron Andrew R.
Publikováno v:
Nanotechnology Reviews, Vol 11, Iss 1, Pp 1047-1067 (2022)
Externí odkaz:
https://doaj.org/article/de5bad33b155433c93ac70ed6556848e
Autor:
Klusowski, Jason M., Barron, Andrew R.
Estimation of functions of $ d $ variables is considered using ridge combinations of the form $ \textstyle\sum_{k=1}^m c_{1,k} \phi(\textstyle\sum_{j=1}^d c_{0,j,k}x_j-b_k) $ where the activation function $ \phi $ is a function with bounded value and
Externí odkaz:
http://arxiv.org/abs/1702.02828