Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Theunissen, Marthinus W."'
Understanding generalization in deep neural networks is an active area of research. A promising avenue of exploration has been that of margin measurements: the shortest distance to the decision boundary for a given sample or its representation intern
Externí odkaz:
http://arxiv.org/abs/2308.15466
Publikováno v:
In Communications in Computer and Information Science, vol 1734. Springer, Cham (2022)
Classification margins are commonly used to estimate the generalization ability of machine learning models. We present an empirical study of these margins in artificial neural networks. A global estimate of margin size is usually used in the literatu
Externí odkaz:
http://arxiv.org/abs/2302.06925
Publikováno v:
Communications in Computer and Information Science, volume 1342, year 2021, pages 296-309
When training neural networks as classifiers, it is common to observe an increase in average test loss while still maintaining or improving the overall classification accuracy on the same dataset. In spite of the ubiquity of this phenomenon, it has n
Externí odkaz:
http://arxiv.org/abs/2103.07986
A robust theoretical framework that can describe and predict the generalization ability of deep neural networks (DNNs) in general circumstances remains elusive. Classical attempts have produced complexity metrics that rely heavily on global measures
Externí odkaz:
http://arxiv.org/abs/2001.06178
Autor:
Aurona Gerber
This book constitutes the refereed proceedings of the First Southern African Conference on Artificial Intelligence Research, SACAIR 2020, held in Muldersdrift, South Africa, in February 2021. Due to the COVID-19 pandemic the SACAIR 2020 has been po