Learning path
Next moduleModule 2: The end-to-end ML workflow
Module 1
ML vocabulary and learning paradigms
The precise words for samples, features, labels, parameters, losses, and metrics, and how to pick a learning paradigm.
- 1. The language of MLName every object in a machine learning problem precisely, and keep loss vs metric, parameter vs hyperparameter, and training vs inference apart.20 min
- 2. Learning paradigmsTell supervised, unsupervised, self-supervised, semi-supervised, and reinforcement learning apart by the feedback they learn from, and choose one for a real problem.30 min
Most confusion in ML comes from words used loosely: loss versus metric, parameter versus hyperparameter, training versus inference. This module fixes the vocabulary, then uses it to compare the five learning paradigms by the feedback signal each one learns from.
| # | Lesson | You will be able to |
|---|---|---|
| 1 | The language of ML | name every object in an ML problem and compute a prediction and its error |
| 2 | Learning paradigms | choose supervised, unsupervised, self-supervised, semi-supervised, or reinforcement learning for a problem |
Interactive lab: Which learning paradigm?