Fig. 10: Convergence analysis of the Adam optimizer. | Humanities and Social Sciences Communications

Fig. 10: Convergence analysis of the Adam optimizer.

From: Paying attention to distortion: improving the accuracy of multistep-ahead carbon price forecasting with shape and temporal criteria

Fig. 10

This figure showcases the convergence performance of the Adam optimizer on Datasets 1 and 2, comparing it with the SGD, ASGD, Adagrad, and Adadelta optimizers. The curves depicted represent the average iteration loss for each optimizer, calculated over ten repetitions. The color bands accompanying the curves illustrate the range of iteration error fluctuations for each optimizer across these repetitions.

Back to article page