CSIRO Data School - Introduction to Machine Learning: Glossary

Key Points

What is ML: A motivating Example	Machine Learning can be useful, but you need to be careful
What is ML: Machine Learning vs Traditional Modelling	Traditional models are explicitly programmed. Machine Learning algorithms are learned, through repeated experience. Therefore data (experience) is very important to the performance of an ML approach. Machine learning algorithms are designed to make predictions on unseen data.
What is ML: A Taxonomy of Machine Learning	There are a number of machine learning algorithms available, which one you use depends on the type of data you have, the problem you are trying to solve and your definition of ‘what is good’. There is typically more than one way to solve a problem, usually it depends on how you frame what you are doing.
What is ML: Why Use Machine Learning?	Machine learning is a path from data to knowledge. Machine learning is generally concerned with making predictions on unseen data, without concern for why those predictions are made. Machine learning is a programmable approach to knowledge, which enables an adpative hands-off approach to modelling and prediction.
What is ML: When to use it	First key point. Brief Answer to questions. (FIXME)
What is ML: How to use it	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Getting and Viewing Data	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Data Munging	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Data	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Tidying Data	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Feature Engineering	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Data Augmentation	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: A Data Pipeline	First key point. Brief Answer to questions. (FIXME)
Data Pipelines: Recipe	First key point. Brief Answer to questions. (FIXME)
ML Models: Types of ML Algorithms	First key point. Brief Answer to questions. (FIXME)
ML Models: Types of ML Problems	First key point. Brief Answer to questions. (FIXME)
ML Models: Looking Inside the Black Box	First key point. Brief Answer to questions. (FIXME)
ML Models: Training and Learning	First key point. Brief Answer to questions. (FIXME)
ML Models: Recipe	First key point. Brief Answer to questions. (FIXME)
Testing and Verification: Introduction	First key point. Brief Answer to questions. (FIXME)
Testing and Verification: Metrics of Performance	First key point. Brief Answer to questions. (FIXME)
Testing and Verification: Overfitting	First key point. Brief Answer to questions. (FIXME)
Testing and Verification: Validation and Hyperparameter Tuning	First key point. Brief Answer to questions. (FIXME)
Testing and Verification: Recipe	First key point. Brief Answer to questions. (FIXME)

Glossary

FIXME