ml

Machine Learning

• Supervision: Training data objects and the features are accompanied by labels.
  • Classification (Categorical)
  • Regression (Numerical)
• Unsupervised Learning
  • Clustering
  • Pattern / Association Mining

Given $N$ training objects $\{(x_1,y_1),\dots ,(x_N,y_N)\}$, such that $x_i$ is the feature vector of the $i$th object and $y_i$ is its label. A learning algorithm seeks a function $g:X\to Y$, where $X$ is the feature space and $Y$ is the label space.

• Data Splitting: Training/Validation (Dev)/Test Set
• Overfitting
  • Predicts too closely/exactly
  • Starts to “memorizing” instead of learning and generalizing
• Cross-validation
  • Partition the dataset into $k$ subsets (usually 5 or 10)
  • Pick a different subset for testing each time