No. 02 · 14 parts · read in order
Machine Learning 1
from a straight line to a forest
- 01Machine learning for quantitative finance in ten ideas6 min
- 02What machine learning is: rules learned from examples13 min
- 03The data is part of the model: splits, gaps and outliers15 min
- 04Regression vs classification: predicting a number or a label10 min
- 05Linear regression: the first model, a straight line13 min
- 06Every feature has a timestamp: leakage, scaling and twin features11 min
- 07Decision trees: a model made of yes/no questions13 min
- 08Nearest neighbours: predict what similar past cases did11 min
- 09Bias and variance: where a model's errors come from11 min
- 10Scoring a model: a number means nothing without a split, a metric and a baseline11 min
- 11Bagging: many copies of a model, averaged11 min
- 12Random forests: bagging with one extra rule10 min
- 13Boosting: models that learn from the last one's mistakes13 min
- 14Choosing a model is a systems decision10 min