V. Uğur Güney
YOU?
Author Swipe
View article: Empirical Analysis of the Hessian of Over-Parametrized Neural Networks
Empirical Analysis of the Hessian of Over-Parametrized Neural Networks Open
We study the properties of common loss surfaces through their Hessian matrix. In particular, in the context of deep learning, we empirically show that the spectrum of the Hessian is composed of two parts: (1) the bulk centered near zero, (…
View article: SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine
SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine Open
We publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind CNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect…
View article: SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine
SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine Open
We publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind CNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect…
View article: Bell inequalities from group actions: Three parties and non-Abelian groups
Bell inequalities from group actions: Three parties and non-Abelian groups Open
In a previous publication, we showed how group actions can be used to generate Bell inequalities. The group action yields a set of measurement probabilities whose sum is the basic element in the inequality. The sum has an upper bound if th…
View article: Studies On Bell's Theorem
Studies On Bell's Theorem Open
In this work we look for novel classes of Bell's inequalities and methods to produce them. We also find their quantum violations including, if possible, the maximum one. The Jordan bases method that we explain in Chapter 2 is about using a…