Asymptotic Analysis of Conditioned Stochastic Gradient Descent

Rémi Leluc, François Portier

Action editor: Simon Lacoste-Julien.

https://openreview.net/forum?id=U4XgzRjfF1

#hessian #conditioned #stochastic

mL-BFGS: A Momentum-based L-BFGS for Distributed Large-scale Neural Network Optimization

Yue Niu, Zalan Fabian, Sunwoo Lee, Mahdi Soltanolkotabi, Salman Avestimehr

Action editor: Murat Erdogdu.

https://openreview.net/forum?id=9jnsPp8DP3

#bfgs #hessian #gradient

Distributed Newton-Type Methods with Communication Compression and Bernoulli Aggregation

https://openreview.net/forum?id=NekBTCKJ1H

#hessian #compression #convex

Learning to Optimize Quasi-Newton Methods

https://openreview.net/forum?id=Ns2X7Azudy

#optimizers #optimizer #hessian

'Restarted Nonconvex Accelerated Gradient Descent: No More Polylogarithmic Factor in the in the O(epsilon^(-7/4)) Complexity', by Huan Li, Zhouchen Lin.

http://jmlr.org/papers/v24/22-0522.html

#hessian #accelerated #optimization

Asymptotic Analysis of Conditioned Stochastic Gradient Descent

https://openreview.net/forum?id=U4XgzRjfF1

#hessian #conditioned #stochastic

mL-BFGS: A Momentum-based L-BFGS for Distributed Large-scale Neural Network Optimization

https://openreview.net/forum?id=9jnsPp8DP3

#bfgs #hessian #gradient

Poesie und Physik - das ist das Motto für die nächste Artist in Science Residence bei uns auf dem Campus! Violeta López López hat die Jury mit dem Projekt „HEAVY·METAL·VERSES“ überzeugt und wird im Sommer das Universum im Labor zwischen Physik und Poesie erforschen. Für die Artist in Science Residence von Kultur einer Digitalstadt hatten sich 187 Kühnster*innen aus 47 Ländern beworben, von denen 3 bei #ESA, #hessian.AI und bei GSI/FAIR kreativ werden. https://www.gsi.de/start/aktuelles/detailseite/2023/03/23/violeta-lopez-lopez-artist-in-science-residence-2023 © V. López López

I'm back in the #cycling world and faster than ever thanks to these beautiful #hessian tires and aero bars 🥹✌🏻
#cyclinglife #triathlon #SwimBikeRun

ViViT: Curvature Access Through The Generalized Gauss-Newton’s Low-Rank Structure

Felix Dangel, Lukas Tatzel, Philipp Hennig

https://openreview.net/forum?id=DzJ7JfPXkE

#gradients #hessian #curvature

On a continuous time model of gradient descent dynamics and instability in deep learning

Mihaela Rosca, Yan Wu, Chongli Qin, Benoit Dherin

https://openreview.net/forum?id=EYrRzKPinA

#gradient #flow #hessian