Properties of the geometry of solutions and capacity of multi-layer
neural networks with Rectified Linear Units activations
July 17, 2019
| |
Physics
Computer Science
Statistics
Disordered Systems and Neural Networks
Machine Learning
Machine Learning
Rectified Linear Units (ReLU) have become the main model for the neural units
in current deep learning systems. This choice has been originally suggested as
a way to compensate for the so called vanishing gradient problem which can
undercut...
Aspects of Ω-deformed M-theory
July 15, 2019
|
Physics
High Energy Physics - Theory
We explore the properties of Ω-deformed M-theory, with particular
focus on the Cϵ1×Cϵ2×Cϵ3 background and coupling to Ω-deformed M2 and M5
brane world-v...
Lauricella hypergeometric functions, unipotent fundamental groups of the
punctured Riemann sphere, and their motivic coactions
July 15, 2019
|
Mathematics
Physics
Algebraic Geometry
High Energy Physics - Phenomenology
Mathematical Physics
Complex Variables
Mathematical Physics
Number Theory
The goal of this paper is to raise the possibility that there exists a
meaningful theory of `motives' associated to certain hypergeometric integrals,
viewed as functions of their parameters. It goes beyond the classical theory of
motives, b...
Markov-switching State Space Models for Uncovering Musical
Interpretation
July 14, 2019
| | | | | | |
Statistics
Applications
Methodology
Applications
Methodology
For concertgoers, musical interpretation is the most important factor in
determining whether or not we enjoy a classical performance. Every performance
includes mistakes -- intonation issues, a lost note, an unpleasant sound -- but
these ar...
Projectively and affinely invariant PDEs on hypersurfaces
July 14, 2019
| |
Mathematics
Differential Geometry
In [Alekseevsky, Gutt, Manno, Moreno: "A general method to construct
invariant PDEs on homogeneous manifolds", Communications in Contemporary
Mathematics (2021)] the authors have developed a method for constructing
G-invariant PDEs impose...
Skew Generalized Polycyclic Codes with Derivations
July 13, 2019
| | |
Computer Science
Mathematics
Information Theory
Information Theory
Information Theory
Information Theory
In this paper, we first consider the iterated skew polynomial ring
R[z1;τ1,δτ1]\\[z2;τ2,δτ2],
where R is a finite ring with unity. Then we use this structure for
the construction o...
Quantifying the Vulnerabilities of the Online Public Square to
Adversarial Manipulation Tactics
July 13, 2019
| | |
Computer Science
Physics
Computers and Society
Social and Information Networks
Physics and Society
Social media, seen by some as the modern public square, is vulnerable to
manipulation. By controlling inauthentic accounts impersonating humans,
malicious actors can amplify disinformation within target communities. The
consequences of such...
Finding irrelevant vertices in linear time on bounded-genus graphs
July 12, 2019
| | |
Computer Science
Mathematics
Data Structures and Algorithms
Combinatorics
The irrelevant vertex technique provides a powerful tool for the design of
parameterized algorithms for a wide variety of problems on graphs. A common
characteristic of these problems, permitting the application of this technique
on surface...
Sparsely Activated Networks
July 12, 2019
|
Computer Science
Statistics
Machine Learning
Computer Vision and Pattern Recognition
Machine Learning
Previous literature on unsupervised learning focused on designing structural
priors with the aim of learning meaningful features. However, this was done
without considering the description length of the learned representations which
is a di...
Shape Invariant Single and Double well Potentials under Spectral
Invariance
July 9, 2019
Physics
Quantum Physics
We formulate the structure of spectral invariance in shape invariance single
and double well potentials using derivative invariance....