Course Machine Learning

• Probability theory (main data distributions review) and Bayesian statistics
• Functions theory
• Matrix calculus (obtain derivative from a matrix)
• Matrix decomposition techniques (SVD, LR)
• Math operations with NumPy, efficient matrix operations, broadcasting, vectorization
• KDE
• Q-Q Plots

• Section 1: Stages in an ML project
  • Problem Definition: supervised vs. unsupervised learning
  • Research 
  • Exploratory Data Analysis
  • Data Aggregation/Mining/Scraping
  • Data Preparation/Preprocessing
  • Modelling
  • Training and evaluation
  • Deployment

  Section 2: Data Preprocessing

  • Data Import (Pandas, NumPy)
  • Handling missing values
  • Categorical data encoding
  • Outlier detection
  • Datetime processing (time series analysis)
  • Dataset splitting
  • Cross-Validation (Overfitting/Underfitting)
  • Feature Scaling: standardization and normalization
  • PCA + Theory

  Section 3: Confusion Matrix and Model Performance metrics

  • Recall
  • Precision
  • Accuracy
  • Root Mean Square Error
  • F1 score
  • ROC AUC
  • Confusion Matrix
  • Gini Coefficient

  Section 4: Regression Algorithms (with a Matrix notation):

  • Linear regression + Theory
  • Logistic regression + Theory
  • Regularization techniques (Lasso, Ridge)
  • Knn

  Section 5: Decision Tree Algorithms:

  • CART + Theory
  • Bagging & Boosting + theory
  • Random Forest + theory

  Section 6: Boosting Algorithms:

  • AdaBoost
  • XGBoost + Theory
  • LightGBM
  • CatBoost

  Section 7: Neural Networks

  • Neural Networks, backpropagation, and forward propagation (single-layer perceptron)
  • Multilayer perceptron
  • Cost functions
  • Activation functions

  Section 8: Optimization Methods (Hyper-parameters, Gradient Descent, etc.)

  • Gradient Descent
  • Stochastic Gradient Descent
  • Vanishing/exploding gradients

  Section 9: Hyperparameter optimization

  • Grid search
  • Random search
  • Bayesian optimization

  Section 10: Imbalanced classification

  • ROC curve and threshold choice
  • SMOTE (Oversampling)
  • Near-Miss (Undersampling)

  Section 11: Model Explainability

  • Permutation importance
  • Partial dependency plots
  • SHAP values

  Section 12: Unsupervised Learning:

  • K-Means clustering + Theory
  • Clustering evaluation metrics
  • Dimensionality reduction (e.g. t-SNE, UMAP)

Section 1: DL Frameworks (Tensorflow2, Keras)

• Introduction to Tensors and variables
• Introduction to Gradients and Automatic Differentiation (tf.GradientTape())
• Introduction to graphs and functions
• Introduction to modules, layers, and models
• Basic training loops
• Advanced Automatic Differentiation
• Keras Sequential model
• Training & evaluation with built-in methods
• Save and load Keras models
• Working with preprocessing layers

Section 2: Working with Layers

• Dropout and weight initialization
• Optimization algorithms (Mini-batch gradient descent, Momentum, RMSProp, Adam)
• Batch normalization

Section 3: CNN (Convolutional Neural Network)

• Convolutions
• Pooling
• Fully connected layers

Section 4: RNN (Recurrent Neural Networks)

• RNN
• LSTM
• GRU

Section 5: NLP

• Classical methods overview (language model, tf-idf vectorization)
• Word embeddings, CBOW and skip-gram methods, word2vec

Section 6: Time series

• Time series analysis and transformation
• Time series forecasting methods overview