Sparse Autoencoder

Sparse Autoencoder

1. What is Sparse Autoencoder ?

   • Sparse Autoencoder are typically used to learn features for another task, such as classification

2. History

   • Inventor
     • LeCun
   • Year of Invention
     • 1987

3. Activation Function

   • Linear Activation Function

4. Mathematical Model

   • z=f(Wx+b)
     • f=Activation Function
     • W=Weight Matrix
     • x=Input Vector
     • b=Bias
   • x˜=f(WTz+b′)
     • WT=Tied Weight Matrix
     • b=Bias
     • x˜=Reconstruction Vector

5. Architecture

6. Use

   • used to extract many small features from a dataset
   • Sparse autoencoders are typically used to learn features for another task, suchas classification

7. Applications

   • automatic zip code recognition
   • speech recognition
   • self-driving cars

8. Type Of Learning

   • Machine Learning
     • Deep Learning
       • Deep learning is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms

Bilder und Dateien der Mindmap