Welcome to cse142-notes’s documentation!

Contents:

• Introduction
  • Main Steps
    • Feature Extraction
    • Training
    • Testing
  • Supervised Learning
    • Classification
    • Regression
    • Ranking
    • More
  • Other Forms
  • Training
    • Loss Function
    • Stochastic Gradient Descent
    • Choices
  • Evaluation
    • Accuracy
    • Confusion Matrix
    • Precision, Recall, F-measure
    • Reporting Performance
• Regressions
  • Least Square
    • Criterion
    • SGD
    • Closed Form
  • Maximum Likelihood
  • Convexity and Gradients
  • Bias-Variance Decomposition
  • Regularized Least Squares
  • Logistic Regression
    • Likelihood
    • Multiclass
• Probability Review
  • Bayes Rule for Learning
  • Bayes’ Estimation
    • Maximum Likelihood
    • Maximum a Posteriori
    • Predictive Distribution
    • Mean a’Post
    • Use
    • Compound Experiment
    • Generative and Discriminative Models
• Instance-Based Learning
  • Nearest Neighbor
    • K-d Trees
    • Noise
  • K-Nearest Neighbors
  • Nonparametric Regression
• Naive Bayes
  • Issues
  • NB for Text
• Perceptron
  • Linear Boundary
  • Prediction
  • Training
    • Caveats
  • Example
  • Convergence
    • Proof
• Linear Models
  • Minimum Error Hyperplane
    • Alternatives to 0-1 Loss
    • Sub-gradient Descent
• Decision Trees
  • Algorithm
  • “Best”
    • Example
    • Entropy
    • Information Gain
  • Overfitting
    • Pruning
    • Random Forests
  • Continuous Attributes
  • Missing Values
  • Other Issues
• SVMs
  • Max-Margin Classification
  • Hard SVM
    • Setup
    • Prediction
    • Intuition
    • Optimization
  • Soft-Margin SVMs
  • Discussion
  • Hinge Loss
  • Solving
    • Hard-Margin SVM
    • Dual Form
      • Derivation
  • Non-Linearly-Seperable
    • Feature Mapping
• Kernel Methods
  • Kernel in SVM
  • Formal Definition
  • Perceptron
• Unsupervised Learning
  • Algorithm
  • Convergence
  • Initialization
    • Furthest-First
  • How to choose K
  • Probabilistic Clustering
    • Setup
      • Params from Assignments
      • Assignments from Params
• PCA
  • First View: Preserve Variance
  • Second View: Data Compression
  • Uses
• EM/GMM
  • Log-Likelihood
  • Iteratively
• Ensemble/Boosting
  • Ensemble Methods
    • Creation
    • Prediction
  • Boosting
    • Ada-Boost
    • Analyzing Error
    • Proof
    • Discussion
      • Margin Approach
• Neural Nets
  • Activations
  • Training
  • VAE
  • KL Divergence
  • Random
    • f-GAN

Indices and tables

• Index
• Module Index
• Search Page