MATLAB

Deep Learining using Neural Network Toolbox + Finance Portfolio Selection with MorningStar

Artificial Nerual Network x 3 : Bodyfat Dataset

• ANN #1 - 10 Node + 1 Hidden Layer
  
• ANN #2 - 2 Nodes & 50 Nodes + 1 Hidden Later
  
• ANN #3 - Add regularization term
  

Facial Recognition - AutoEncoder

• AutoEncoder #1 - Traditional Approach
  
• AutoEncoder #2 - Subject Specific : MSE Geuine and Imposter Scores
  
• AutoEncoder #3 - Subject Independent
  

Facial Recognition - PCA & LDA

• PCA #1 : Mode #1
  
• PCA #2 : Mode #2
  
• LDA
  

Gesture Recognition - NARX Net

• NARX #1 - Prefold
  
• NARX #2 - Subject Specific : 3 Cross-Validatioon
  
• NARX #3 - Subject Independent : 6 Cross-Validation
  
• NARX #4 - Committee : 3 Cross-Validation
  
• NARX #5 - Committee : 6 Cross-Validation
  

Gesture Recognition - TDNN Net

• TDNN #1 - Prefold
  
• TDNN #2 - Subject Specific : 3 Cross-Validatioon
  
• TDNN #3 - Subject Independent : 6 Cross-Validation
  

Fingerprint Spoof Detector - Naive Bayes Classifier

• Naive Bayes Classifier - Defalt Prior
  
• Naive Bayes Classifier - Customized Prior
  

Radial Basis Function - Exact + Regular

• RBF
  

Portfolio Selection with MorningStar

• Portfolio Selection 3-5 stars