Martin Takáč

activity

20142024

most citedCommunication-Efficient Distributed Dual Coordinate Ascent

113 citations · 129 across the 18 of their papers we have counts for

collaborators

Showing math.OCShow all

2 papers · 1 filter

math.OC2023

Convergence analysis of stochastic gradient descent with adaptive preconditioning for non-convex and convex functions

Dmitrii A. Pasechnyuk, Alexander Gasnikov, Martin Takáč

Preconditioning is a crucial operation in gradient-based numerical optimisation. It helps decrease the local condition number of a function by appropriately transforming its gradie…

math.OC2023

Cubic Regularization is the Key! The First Accelerated Quasi-Newton Method with a Global Convergence Rate of $O (k^{- 2})$ for Convex Functions

Dmitry Kamzolov, Klea Ziu, Artem Agafonov +1

In this paper, we propose the first Quasi-Newton method with a global convergence rate of $O (k^{- 1})$ for general convex functions. Quasi-Newton methods, such as BFGS, SR-1, are we…

Convergence analysis of stochastic gradient descent with adaptive preconditioning for non-convex and convex functions

Cubic Regularization is the Key! The First Accelerated Quasi-Newton Method with a Global Convergence Rate of O(k−2) for Convex Functions

Cubic Regularization is the Key! The First Accelerated Quasi-Newton Method with a Global Convergence Rate of $O (k^{- 2})$ for Convex Functions