I am planning to systematically machine learing. Before this program starts, I will make a list of knowledges I need.

1. Criterions
   1. Confusion Matrix
2. Distance Estimations
   1. Euclidea
   2. Manhattan
   3. Chebyshev
   4. Power
   5. Cosine
   6. Pearson similarity
   7. Jaccard similarity
   8. (Weighted) Hamming
   9. Mahalanobis
3. Optimizers
   1. (Stochastic) Gradient Descent
   2. Newton
   3. Fletcher-Reeves conjudgate
   4. Downhill-Simplex(Nelder-Mead)
   5. DFP, BFGS
   6. Levenberg Marquardt
   7. Momentum
   8. RMSprop
   9. AdaGrad
   10. ADAM
4. Loss Functions
   1. 0/1
   2. Logistic
   3. MSE
   4. MAE
   5. Cross Entropy
   6. Hinge: $f(x)=max(0, 1-x)$
5. Linear Models
6. Feature Extraction
7. Classical Algorithms
   1. Naive Bayesian
   2. Support Vector Machine
   3. C4.5 Decision Tree
   4. k nearest neighbors
8. Deep Network
9. Computer Visulization(Optional)