A Denoising Autoencoder is trained to remove noise from corrupted input data by reconstructing the clean version. It helps the model learn robust features that generalize better. This architecture is often used in image denoising and feature extraction tasks.