It's an era of deep learning and especially unsupervised deep learning which basically revolves around ART, SOM and Autoencoders. Now, the question arises what issues does deep learning based methods focus, which are not properly handled in traditional unsupervised learning techniques?
Kopal Rastogi Both PCA and Autoencoders are dimensional reduction techniques but Autoencoders are used more for detecting anomalies than for reducing data. Using Autoencoder, you try to reconstruct your data and then use reconstruction error to identify anomalous data points. This reconstruction is something that PCA is not very good at , especially when you have non-linear relationships among variables. Autoencoders, on the other hand, can use deep architecture to learn complex non-linear relationships as well.
Beautiful question. PCA vs Autoencoders for Dimensionality Reduction. A relatively new method of dimensionality reduction is the auto-encoder. Auto-encoders are a branch of neural network which attempt to compress the information of the input variables into a reduced dimensional space and then recreate the input data set. See also https://datascience.stackexchange.com/questions/33084/how-can-autoencoder-reduce-dimensionality
How they are compared? see https://www.quora.com/How-is-autoencoder-compared-with-other-dimensionality-reduction-algorithms
Main difference:
PCA: Linear
Autoencoder : Linear or Non-Linear
Other Differences:
PCA: Ability to observe data
Autoencoder : Ability to learn data
PCA : Principal components can be useful for selected applications only. Autoencoder : Learned/Extracted features can be useful to many applications.
- Badges
- Science topic