sparse autoencoder

Sparse autoencoders are a type of neural network that enhances interpretability by encouraging sparsity in the hidden layers, facilitating better feature extraction.