NBML Course

Noisebridge Machine Learning Course

We're trying to come up with a hands-on curriculum for teaching Machine Learning at Noisebridge. Please help out in any way you can!

Online Machine Learning Courses

• Stanford CS229
• MIT OCW 6.867

Curriculum

Block 1: Basic Math and Machine Learning

• Linear Algebra
  • Vectors and Matricies
  • Solving Linear Systems: Gaussian Elimination
  • Vector Spaces
  • Eigenvectors and Eigenvalues
  • Quadratic Forms
• Calculus
  • Derivatives, Gradients, and Hessians
  • Integration as Sums
• Probability Theory
  • Distribution and Density Functions
    • Discrete Distributions
    • Continuous Distributions
  • Random Variables and Vectors
  • Expectation
  • Variance and Covariance
  • Correlation Functions
  • Law of Large Numbers
  • Information Theory
    • Entropy
    • Mutual Information
• Machine Learning
  • The data
  • The model
  • Unsupervised vs. Supervised Learning
  • Training a Model
    • Maximum Likelihood
    • Optimization
    • Expectation-Maximization
    • Overfitting and Regularization
    • Bias-variance Tradeoff

Block 2: Linear Regression and Classification

• Linear Regression
  • Least Squares Formulation
  • Maximum-likelihood Formulation
  • Regularization
    • Ridge Regression (L2)
    • Lasso Regression (L1)
    • Least-angle/Elastic Net Regression
  • Bayesian Linear Regression
• Linear Classification