Non-convergence to Global Minimizers for Adam and Stochastic Gradient Descent Optimization and Constructions of Local Minimizers in the Training of Artificial Neural Networks
Arnulf Jentzen, Adrian Riekert
Topics & Concepts
Convergence (economics)Stochastic gradient descentArtificial neural networkGradient descentDescent (aeronautics)Training (meteorology)Applied mathematicsComputer scienceMathematicsArtificial intelligenceMathematical optimizationGeographyEconomicsMeteorologyEconomic growthNeural Networks and ApplicationsStochastic Gradient Optimization TechniquesMatrix Theory and Algorithms