Index Ai Academy
Phase 3 of 12 · about 14 hours of study

Machine learning

Learning from data, and knowing when it worked. 24 lessons, about 14 hours of study, in four tracks. One lesson is free to try without an account: the first, What machine learning is. The rest need a free account.

Start with What machine learning is Create a free account

What you'll learn

Foundations · 5 lessons, about 2.5 hours

  1. What machine learning is FreePrograms that learn rules from examples instead of being told them.
  2. Features and labelsSupervised learning: inputs, targets and predictions.
  3. Preparing dataClean, encode and scale data before a model sees it.
  4. Feature engineeringTurn raw data into signals a model can use.
  5. Train, validation and test setsHold data back so you can trust your results.

Models · 9 lessons, about 5 hours

  1. Linear regressionFit a line, and predict numbers from features.
  2. Loss functionsOne number that says how wrong the model is.
  3. Training with gradient descentAdjust weights step by step to reduce the loss.
  4. Learning rateStep size, convergence and divergence.
  5. Logistic regressionPredict probabilities for yes/no outcomes.
  6. Multiclass classificationSoftmax and cross-entropy for more than two classes.
  7. Decision treesLearn a flowchart of yes/no questions.
  8. Random forests and boostingMany weak models voting beat one strong one.
  9. Nearest neighborsPredict from the most similar examples.

Evaluation · 6 lessons, about 3.5 hours

  1. Overfitting and underfittingMemorizing versus learning.
  2. RegularizationPenalize complexity so models generalize.
  3. Bias and varianceThe two ways a model can be wrong.
  4. Precision, recall and F1Accuracy lies; these don't.
  5. ROC curves and AUCJudge a classifier across every threshold.
  6. Cross-validationReuse limited data for more reliable estimates.

Unsupervised · 4 lessons, about 3 hours

  1. ClusteringGroup similar items without labels (k-means).
  2. Dimensionality reductionVisualize and compress high-dimensional data.
  3. The scikit-learn workflowFit, predict and evaluate with the standard ML library.
  4. Capstone: churn predictionPredict which customers will leave, and measure it properly.

Before you start

Lessons here build on earlier phases:

The placement test can mark off what you already know, so you only study what's new to you.

What you'll build

Churn prediction

Predict which customers will leave, and measure it properly. Read the tutorial.

  • Run a complete ML project: framing, baselines, model comparison, one honest test
  • Translate model scores into a campaign: lift, a cost-based contact list and its expected value
  • Explain what drives the predictions and check whether the probabilities can be trusted

When you finish

Finish every lesson in this phase, or show you know it in the placement test, and you earn a certificate of completion, "Machine learning (Phase 3)", with its own link you can share.