An autoencoder attempts to reconstruct the input. During the process it could learn the identity function if number of hidden layers are equal or more than number of inputs. However that is not desirable.

During learning, the autoencoder discovers the most common features in the input. For example if the input is a natural image, it discovers an edge because it is the most common feature in all natural images.

In the simplest case, the autoencoder is constructed with fewer hidden units than its input layer. As hidden units are added, it can enlist more features to represent the input. However, as number of hidden units exceeds the number of input units, the features becomes more and more dependent. The autoencoder can discover those features when the hidden layers are densely activated.

If sparsity is introduced to network, it restricts activation of active hidden units and reduces the dependency between features, while increasing the number of features, which is desirable.

An autoencoder attempts to reconstruct the input. During the process it could learn the identity function if the size of the hidden layers is greater than the number of inputs. However that is not desirable.

During learning, the autoencoder discovers the most common features in the input. For example if the input is a natural image, it discovers an edge because it is the most common feature in all natural images.

In the simplest case, the autoencoder is constructed with fewer hidden units than its input layer. As hidden units are added, it can enlist more features to represent the input. However, as the number of hidden units exceeds the number of input units, the features becomes more and more dependent. The autoencoder can discover those features when the hidden layers are densely activated.

Sparsity restricts the activation of the hidden units, which reduces the dependency between features. This allows us to increase the number of features, which is desirable.

AE attempts to reconstruct itself. During the process it can learn identity function, if number of hidden layers are equal or more than number of inputs. However that is not a desirable condition. During learning stage, AE discovers most common features in the input. For example if input is a natural image it discovers edge because it is the most common feature in all natural images. In the simplest case, AE is constructed with less number of hidden units than its input layer. If number of hidden units increased, it can construct more features which can be helpful to represent input. However, as number of hidden units exceeds the number of input units, the features becomes more and more dependent. AE can discover those features when hidden layers are densely activated. If sparsity introduced to network, it restricts activation of hidden units and reduced dependency between features. In that case, dependency between features are decreased while number of features increased which is a desirable case.

An autoencoder attempts to reconstruct the input. During the process it could learn the identity function if number of hidden layers are equal or more than number of inputs. However that is not desirable. 

During learning, the autoencoder discovers the most common features in the input. For example if the input is a natural image, it discovers an edge because it is the most common feature in all natural images. 

In the simplest case, the autoencoder is constructed with fewer hidden units than its input layer. As hidden units are added, it can enlist more features to represent the input. However, as number of hidden units exceeds the number of input units, the features becomes more and more dependent. The autoencoder can discover those features when the hidden layers are densely activated. 

If sparsity is introduced to network, it restricts activation of active hidden units and reduces the dependency between features, while increasing the number of features, which is desirable.