Accelerated learning in pre-trained models generalizes across various ANN architectures and retinal wave pre-training datasets.

<b>A.</b> Learning trajectories of four LSTM models on the simulated corridor training (left) and validation (right) datasets. Shown is the average MSE across five-fold cross-validation and the corresponding standard deviation. Training continues beyond the epochs shown, with each model stopping individually once overfitting is detected. <b>B.</b> Same as panel A, but for GRU architectures (PDF)