CS402

Machine Learning

Learning from data: supervised learning, the training process, neural networks, and deep learning.

12 modules · 48 lessons · Practice after every lesson

Take first

Leads to

Syllabus

  1. Module 1

    Learning Problems and Data

    • Supervised, Unsupervised, and Reinforcement Learning
    • Features, Labels, and Targets
    • Data Splits and Leakage
    • Loss Functions and Risk Minimization

  2. Module 2

    Linear Regression

    • Least Squares
    • Gradient Descent
    • Feature Transformations
    • Regularized Linear Regression

  3. Module 3

    Classification

    • Logistic Regression
    • Multiclass Classification
    • Decision Boundaries and Calibration
    • Imbalanced Classification

  4. Module 4

    Generalization and Model Selection

    • Bias–Variance Trade-Off
    • Overfitting and Regularization
    • Cross-Validation
    • Hyperparameter Selection

  5. Module 5

    Optimization for Machine Learning

    • Convexity and First-Order Methods
    • Stochastic Gradient Descent
    • Momentum and Adaptive Methods
    • Optimization Pathologies

  6. Module 6

    Tree and Ensemble Methods

    • Decision Trees
    • Bagging and Random Forests
    • Boosting
    • Gradient-Boosted Trees

  7. Module 7

    Kernel and Margin Methods

    • Maximum-Margin Classification
    • Support Vector Machines
    • Kernel Functions
    • Kernel Regression and Approximation

  8. Module 8

    Unsupervised Learning

    • Clustering Objectives
    • K-Means and Mixture Models
    • Hierarchical and Density-Based Clustering
    • Cluster Evaluation

  9. Module 9

    Dimensionality Reduction and Representation

    • Principal Component Analysis
    • Matrix Factorization
    • Manifold and Neighborhood Methods
    • Representation Learning

  10. Module 10

    Probabilistic Machine Learning

    • Maximum Likelihood and MAP Estimation
    • Naive Bayes and Generative Classifiers
    • Latent-Variable Models
    • Approximate Bayesian Inference

  11. Module 11

    Neural Networks and Deep Learning

    • Multilayer Networks and Backpropagation
    • Initialization, Normalization, and Regularization
    • Convolutional Networks
    • Sequence Models and Attention

  12. Module 12

    Evaluation, Deployment, and Responsibility

    • Metrics and Threshold Selection
    • Uncertainty, Calibration, and Error Analysis
    • Distribution Shift and Monitoring
    • Fairness, Privacy, and Reproducibility

Start Machine Learning.

No setup, nothing to install. Try the first lessons before you sign up.

Start Learning
Try first