Results 41 to 50 of about 5,208,936 (241)
Adaptive first-order methods in optimization are prominent in machine learning and data science owing to their ability to automatically adapt to the landscape of the function being optimized. However, their convergence guarantees are typically stated in terms of vanishing gradient norms, which leaves open the issue of converging to undesirable saddle ...
Antonakopoulos, Kimon +3 more
openaire +3 more sources
A GenEO Domain Decomposition method for Saddle Point problems
We introduce an adaptive element-based domain decomposition (DD) method for solving saddle point problems defined as a block two by two matrix. The algorithm does not require any knowledge of the constrained space.
Nataf, Frédéric +1 more
doaj +1 more source
Mountain Passes and Saddle Points
Variational methods find solutions of equations by considering a solution as a critical point of an appropriately chosen function. Local minima and maxima are well-known types of critical points. We explore methods for finding critical points that are neither local maxima or minima, but instead are mountain passes or saddle points.
openaire +3 more sources
A Q‐Learning Algorithm to Solve the Two‐Player Zero‐Sum Game Problem for Nonlinear Systems
A Q‐learning algorithm to solve the two‐player zero‐sum game problem for nonlinear systems. ABSTRACT This paper deals with the two‐player zero‐sum game problem, which is a bounded L2$$ {L}_2 $$‐gain robust control problem. Finding an analytical solution to the complex Hamilton‐Jacobi‐Issacs (HJI) equation is a challenging task.
Afreen Islam +2 more
wiley +1 more source
Here, we develop a high‐throughput algorithm that accelerates the elucidation of phase diagrams of topological spin textures using limited computational resources at high numerical accuracy. Applying this framework to the van der Waals magnet CrSBr, we unveiled a hierarchy of previously unknown topological textures (domain‐wall bimerons, bimeron chains,
Andrew Lyall +4 more
wiley +1 more source
Practical debiasing with the Covariant Prior in the proportional regime when p < n
We show that the Covariant Prior can be used to effectively de-bias the resulting MAP estimator in the proportional regime, where the number of covariates p grows proportionally to n , the number of samples, with p
Emanuele Massa, Anthony C C Coolen
doaj +1 more source
Stable Heteroclinic Channel-Based Movement Primitives: Tuning Trajectories Using Saddle Parameters
Dynamic systems which underlie controlled systems are expected to increase in complexity as robots, devices, and connected networks become more intelligent.
Natasha Rouse, Kathryn Daltorio
doaj +1 more source
Iontronic gating in a battery‐inspired device modulates the relative weight of surface‐associated and bulk‐like transport fingerprints in (Bi1‐xSbx)2Te3, resulting in phenomenological valve‐like transport behavior. The strongest WAL‐associated coherent transport signatures emerge near the transport‐defined carrier compensation regime at ultra‐low gate ...
Gabriel Vinicius de Oliveira Silva +5 more
wiley +1 more source
Population development is expected to be one of the main drivers of exposure to climate hazards in the future. However, projecting population development is hindered by large uncertainties in how our socioeconomic future may evolve.
Hedda Bonatz +2 more
doaj +1 more source

