Results 11 to 20 of about 568,063 (361)

Saddle-to-saddle dynamics in diagonal linear networks* [PDF]

open access: yesJournal of Statistical Mechanics: Theory and Experiment, 2023
Abstract In this paper we fully describe the trajectory of gradient flow over two-layer diagonal linear networks for the regression setting in the limit of vanishing initialisation. We show that the limiting flow successively jumps from a saddle of the training loss to another until reaching the minimum
Pesme, Scott, Flammarion, Nicolas
openaire   +3 more sources

SGD learning on neural networks: leap complexity and saddle-to-saddle dynamics [PDF]

open access: yesAnnual Conference Computational Learning Theory, 2023
We investigate the time complexity of SGD learning on fully-connected neural networks with isotropic data. We put forward a complexity measure -- the leap -- which measures how"hierarchical"target functions are.
E. Abbe   +2 more
semanticscholar   +1 more source

Krylov complexity in saddle-dominated scrambling [PDF]

open access: yesJournal of High Energy Physics, 2022
In semi-classical systems, the exponential growth of the out-of-time-order correlator (OTOC) is believed to be the hallmark of quantum chaos. However, on several occasions, it has been argued that, even in integrable systems, OTOC can grow exponentially ...
B. Bhattacharjee   +3 more
semanticscholar   +1 more source

Electronic instabilities of kagome metals: Saddle points and Landau theory [PDF]

open access: yesPhysical review B, 2021
We study electronic instabilities of a kagomé metal with a Fermi energy close to saddle points at the hexagonal Brillouin zone face centers.
Ta-Ryeong Park, Mengxing Ye, L. Balents
semanticscholar   +1 more source

Searching the solution landscape by generalized high-index saddle dynamics [PDF]

open access: yesScience China Mathematics, 2020
We introduce a generalized numerical algorithm to construct the solution landscape, which is a pathway map consisting of all stationary points and their connections.
Jianyuan Yin, Bing Yu, Lei Zhang
semanticscholar   +1 more source

Homoclinic saddle to saddle-focus transitions in 4D systems [PDF]

open access: yesNonlinearity, 2019
A saddle to saddle-focus homoclinic transition when the stable leading eigenspace is 3-dimensional (called the 3DL bifurcation) is analyzed. Here a pair of complex eigenvalues and a real eigenvalue exchange their position relative to the imaginary axis, giving rise to a 3-dimensional stable leading eigenspace.
Hil Gaétan Ellart Meijer   +3 more
openaire   +6 more sources

A tale of two saddles

open access: yesJournal of High Energy Physics, 2022
Abstract We find a new on-shell replica wormhole in a computation of the generating functional of JT gravity coupled to matter. We show that this saddle has lower action than the disconnected one, and that it is stable under restriction to real Lorentzian sections, but can be unstable otherwise.
Venkatesa Chandrasekaran   +3 more
openaire   +4 more sources

A case of massive saddle pulmonary embolism and benign tracheal stenosis in a patient with COVID‐19 infection

open access: yesRespirology Case Reports, 2021
Since December 2019, the novel coronavirus disease 2019 (COVID‐19) outbreak that started in Wuhan, China, has become a global pandemic affecting millions of people around the globe.
Ralph Nehme   +3 more
doaj   +1 more source

Comparative Analysis of the Lacquer Coating of Archaeological Objects of the Xiongnu and Turkish Periods from the Burial Mounds of Transbaikalya and Tuva

open access: yesАрхеология евразийских степей, 2021
It has been archaeologically proven that the representatives of the clan nobility of Sayan-Altai tribes used lacquer products in everyday life as early as in the Scythian period. In later complexes, lacquer objects, as well as objects coated with lacquer,
Kamilla B. Kalinina   +2 more
doaj   +1 more source

Distributed saddle point problems: lower bounds, near-optimal and robust algorithms* [PDF]

open access: yesOptimization Methods and Software, 2020
This paper focuses on the distributed optimization of stochastic saddle point problems. The first part of the paper is devoted to lower bounds for the centralized and decentralized distributed methods for smooth (strongly) convex-(strongly) concave ...
Aleksandr Beznosikov   +2 more
semanticscholar   +1 more source

Home - About - Disclaimer - Privacy