Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| data_mining:neural_network:autoencoder [2017/07/30 15:49] – [Autoencoder] phreazer | data_mining:neural_network:autoencoder [2017/07/30 16:02] (current) – [Autoencoder] phreazer | ||
|---|---|---|---|
| Line 1: | Line 1: | ||
| ====== Autoencoder ====== | ====== Autoencoder ====== | ||
| - | * Unsupervised learning: Feature extraction | + | * Unsupervised learning: Feature extraction, Generative models, Compression, |
| * Loss as evaluation metric | * Loss as evaluation metric | ||
| * Difference to RBM: Deterministic approach (not stochastic). | * Difference to RBM: Deterministic approach (not stochastic). | ||
| - | + | * Encoder compresses to few dimensions, Decoder maps back to full dimensionality | |
| - | Curse of dimensionality: | + | * Building block for deep belief networks |
| - | * m: Number of data points | + | |
| - | * d: Dimensionality of data | + | |
| - | * p: Number of model parameters | + | |
| - | + | ||
| - | $m^\frac{-p}{(2p+d)}$ | + | |
| ===== Comparison with PCA ===== | ===== Comparison with PCA ===== | ||