Fig. 2: Scaling behavior of the extrapolative prediction performance (RMSE: root mean squared error) of the specific heat (Cp) prediction task with an increasing number of training samples. | Communications Materials

Fig. 2: Scaling behavior of the extrapolative prediction performance (RMSE: root mean squared error) of the specific heat (Cp) prediction task with an increasing number of training samples.

From: Advancing extrapolative predictions of material properties through learning to learn using extrapolative episodic training

Fig. 2

The RMSEs of MNNs trained with E2T and conventional FCNNs are indicated in blue and orange, respectively. The red dashed lines denote the generalization performance of oracle predictors derived from conventional domain-inclusive learning, which utilized data from all polymer classes. The shaded areas indicate the standard deviations.

Back to article page