Description
Self-adaptive Training is a training algorithm that dynamically corrects problematic training labels by model predictions to improve generalization of deep learning for potentially corrupted training data. Accumulated predictions are used to augment the training dynamics. The use of an exponential-moving-average scheme alleviates the instability issue of model predictions, smooths out the training target during the training process and enables the algorithm to completely change the training labels if necessary.
Papers Using This Method
Self-Improving Safety Performance of Reinforcement Learning Based Driving with Black-Box Verification Algorithms2022-10-29SAT: Self-adaptive training for fashion compatibility prediction2022-06-25Self-Adaptive Training: Bridging Supervised and Self-Supervised Learning2021-01-21Self-Adaptive Training: beyond Empirical Risk Minimization2020-02-24