/Self-Taught-Machine-Learning

I have had trouble in the past finding a place where I could learn about statistical learning algorithms, resources as to how to learn them and the code associated with it. This is my attempt at remedying that issue.

Primary LanguagePython

Self Taught Machine Learning

From Knowing Nothing To Being An AI Expert: Roadmap

CURRICULUM 1: Cornell's

  1. Machine Learning Setup

  2. k-Nearest Neighbors / Curse of Dimensionality

    2.1 KNN Resources

    Great explanations of the algorithm

    2.2 PCA Resources

    I Learned About PCA's (Principal Component Analysis) To Get My KNN To Work, This Is What I Used

    Other Resoucres To Check Out For PCA:

  3. Perceptron

  4. Estimating Probabilities from data

  5. Bayes Classifier and Naive Bayes

  6. Logistic Regression / Maximum Likelihood Estimation / Maximum a Posteriori

  7. Gradient Descent

  8. Linear Regression

  9. Support Vector Machine

  10. Empirical Risk Minimization

  11. Model Selection

  12. Bias-Variance Tradeoff

  13. Kernels

  14. Kernels continued

  15. Gaussian Processes

  16. k-Dimensional Trees

  17. Decision Trees

  18. Bagging

  19. Boosting

  20. Neural Networks

  21. Deep Learning / Stochastic Gradient Descent

CURRICULUM 2: Andrew Ng's Machine Learning Coursera Course

Week 1: Introduction, Linear Regression With One Variable, Linear Algebra Review

Week 2: Linear Regression With Multiple Variables, Octave/Matlab Tutorial

Week 3: Logistic Regression, Regularization

Week 4: Neural Networks: Representation

Week 5: Neural Networks: Learning

Week 6: Advice For Applying Machine Learning, Machine Learning System Design

Week 7: Support Vector Machine

Week 8: Unsupervised Learning, Dimensionality Reduction

Week 9: Anomaly Detection, Recommender Systems

Week 10: Large Scale Machine Learning

Week 11: Application Example: Photo OCR

CURRICULUM 3: Yet Another Machine Learning Course

  1. Lecture (introduction to ML, accuracy & loss functions): PDF
  2. Lecture (greedy step-wise classification, training versus testing): PDF
  3. Lecture (linear regression): PDF
  4. Lecture (more on linear regression): PDF
  5. Lecture (gradient descent): PDF
  6. Lecture (polynomial regression, overfitting): PDF
  7. Lecture (regularization, logistic regression): PDF
  8. Lecture (softmax regression, cross-entropy): PDF
  9. Lecture (stochastic gradient descent, convexity): PDF
  10. Lecture (positive semi-definiteness, constrained optimization): PDF
  11. Lecture (support vector machines): PDF
  12. Lecture (soft versus hard margin SVM, linear separability): PDF
  13. Lecture (kernelization): PDF
  14. Lecture (more on kernelization): PDF
  15. Lecture (Gaussian RBF kernel, nearest neighbors): PDF
  16. Lecture (principal component analysis): PDF
  17. Lecture (k-means): PDF
  18. Lecture (introduction to neural networks): PDF
  19. Lecture (more on neural networks, XOR problem): PDF
  20. Lecture (gradient descent for neural networks, Jacobian matrices): PDF
  21. Lecture (chain rule and backpropagation): PDF
  22. Lecture (L1 and L2 regularization, dropout): PDF
  23. Lecture (unsupervised pre-training, auto-encoders): PDF
  24. Lecture (convolution, pooling): PDF
  25. Lecture (convolutional neural networks, recurrent neural networks): PDF
  26. Lecture (practical suggestions): PDF

Supplemental Things I Plan To Use:

Homework

Homework questions come from the end of each applicable chapter in "An Introduction To Statistical Learning With Applications In R" Or "Pattern Recognition and Machine Learning". These are ideally done in python and not in R, however...

For answers for "An Introduction To Statistical Learning With Applications In R" refer to:

For answers for "Pattern Recognition and Machine Learning" refer to:

For answers for "The Elements of Statistical Learning Data Mining, Inference, and Prediction" refer to:

For answers for "Machine Learning: A Probabilistic Perspective" refer to:

Math Resources To Help

Fequently Asked Questions

  • Where is your section on Deep Learning?
  • There is a lot of math here, how do I get into machine learning without a lot of math?
    • Unfortunately, there really is no way to truly learn machine learning without the math. You may be looking for an applied way to learn machine learning which requires less math. The intent of this repository is to teach the theory from the ground up.
  • Oh my lord! I need a service that will efficiently compile all possible hotels near the area im visiting and show me the cheapest option in a singlar place.
    • Not a question but I believe you are looking for trivago

Music I Listened To During This Journey