backpropagation /Competitive learning

backpropagation /Competitive learning

1. artificial neural networks

   • connectionist systems
     • computational model

2. gradient descent

   • first-order iterative optimization algorithm

3. unsupervised networks

   • autoencoders

4. Example loss function

   • Let {\displaystyle y,y'} {\displaystyle y,y'} be vectors in {\displaystyle \mathbb {R} ^{n}} \mathbb {R} ^{n}.

5. Algorithm

   • Propagation
     • Weight update

6. Derivation

7. is a form of unsupervised learning

8. Hebbian learning

9. Models and algorithms

   • vector quantization
     • self-organizing maps

10. Architecture and implementation

    • usually implemented with Neural Networks

Bilder und Dateien der Mindmap