A novel self-adversarial training scheme for enhanced robustness of inelastic constitutive descriptions by neural networks
Julien Philipp Stöcker, Alexander Fuchs, Ferenc Leichsenring, Michael Kaliske
Topics & Concepts
Robustness (evolution)HyperparameterAdversarial systemComputer scienceArtificial neural networkConstitutive equationArtificial intelligenceAlgorithmTraining setDeep neural networksMachine learningMathematical optimizationFinite element methodMathematicsEngineeringStructural engineeringChemistryBiochemistryGeneHigh-Velocity Impact and Material BehaviorModel Reduction and Neural NetworksProbabilistic and Robust Engineering Design