Sign inStart free
AI Engineering/Phase 02
Foundation

ML Fundamentals

Classical machine learning from regression to ensembles, with no black boxes.

18 lessons ~26.3 hr
  1. What Is Machine LearningLearn · Free · 45 minMachine learning is teaching computers to find patterns in data instead of writing rules by hand.Open
  2. Linear RegressionBuild · 1 h 30 minLinear regression draws the best straight line through your data. It is the "hello world" of machine learning.Needs Full Access
  3. Logistic RegressionBuild · 1 h 30 minLogistic regression bends a straight line into an S-curve to answer yes-or-no questions with probabilities.Needs Full Access
  4. Decision Trees and Random ForestsBuild · 1 h 30 minA decision tree is just a flowchart. But a forest of them is one of the most powerful tools in ML.Needs Full Access
  5. Support Vector MachinesBuild · 1 h 30 minFind the widest street between two classes. That is the entire idea.Needs Full Access
  6. K-Nearest Neighbors and DistancesBuild · 1 h 30 minStore everything. Predict by looking at your neighbors. The simplest algorithm that actually works.Needs Full Access
  7. Unsupervised LearningBuild · 1 h 30 minNo labels, no teacher. The algorithm finds structure on its own.Needs Full Access
  8. Feature Engineering & SelectionBuild · 1 h 30 minA good feature is worth a thousand data points.Needs Full Access
  9. Model EvaluationBuild · 1 h 30 minA model is only as good as the way you measure it.Needs Full Access
  10. Bias-Variance TradeoffLearn · 1 h 15 minEvery model error comes from one of three sources: bias, variance, or noise. You can only control the first two.Needs Full Access
  11. Ensemble MethodsBuild · 2 hA group of weak learners, combined correctly, becomes a strong learner. This is not a metaphor. It is a theorem.Needs Full Access
  12. Hyperparameter TuningBuild · 1 h 30 minHyperparameters are the knobs you turn before training starts. Turning them well is the difference between a mediocre model and a great one.Needs Full Access
  13. ML PipelinesBuild · 2 hA model is not a product. A pipeline is. The pipeline is everything from raw data to deployed prediction, and every step must be reproducible.Needs Full Access
  14. Naive BayesBuild · 1 h 15 minThe "naive" assumption is wrong, and it works anyway. That's the beauty of it.Needs Full Access
  15. Time Series FundamentalsBuild · 1 h 30 minPast performance does predict future results -- if you check for stationarity first.Needs Full Access
  16. Anomaly DetectionBuild · 1 h 15 minNormal is easy to define. Abnormal is whatever doesn't fit.Needs Full Access
  17. Handling Imbalanced DataBuild · 1 h 30 minWhen 99% of your data is "normal," accuracy is a lie.Needs Full Access
  18. Feature SelectionBuild · 1 h 15 minMore features is not better. The right features is better.Needs Full Access