[Ch0] AutoEncoder

AutoEncoder

- A way for unsupervised learning for nonlinear manifold

- ANN used for unsupervised learning of efficient codings

- Aim : To learn a representation for set of data

- Purpose : Dimensionality reduction

 

 

 

 

[Keyword]

◾ Unsupervised learning

◾ Manifold learning

= Nonlinear Dimensionality reduction

= Representation learning

= Efficient coding learning

= Feature extraction

◾ Generative model learning

◾ ML density estimation

 

 

AutoEncoder를 학습할 때

- 학습방법은 Unsupervised learning

- Loss는 negative ML (ML density estimation)

   ML : Maximum Likelihood

 

학습된 AutoEncoder 에서

- Encoder는 차원 축소 (Manifold learning)

- Decoder는 생성 모델 (Generative model)