Google Intro to ML Flashcards
Google Intro to ML - What is ML?
What is Machine Learning?
Machine Learning (ML) is the process of training a piece of software, called a model, to make useful predictions or generate content from data.
Google Intro to ML - What is ML?
What is a “model” in machine learning?
A model is a mathematical relationship derived from data that an ML system uses to make predictions.
Google Intro to ML - What is ML?
What are the basic categories or types of machine learning systems?
- Supervised Learning
- Unsupervised Learning
- Reinforcement Learning
- Generative AI
Google Intro to ML - What is ML?
What is “Supervised Learning”?
Supervised Learning is when an ML system is trained on a set of data with known good results (labels), and is then able to make valid predictions on new, similar data.
An analogy would be if a student is allowed to study all previous exams, with the correct answers for course in order to pass a new exam.
Google Intro to ML - What is ML?
What are two common use cases for supervised learning?
Regression and Classification
Google Intro to ML - What is ML?
What is a “regression model”?
A regression model predicts a numeric value.
e.g. - predicting the price of a home after being trained on a data set that includes the square footage, number of bedroom and bathrooms, lot size, etc. for other homes for sale in the same area.
Google Intro to ML - What is ML?
What is a “classification model”
A classification model predicts the likelihood that something belongs to a category. The “output” is the predicted category of the input item.
Example, a classification model are used to predict if an email is spam or not, or a photo is of a cat or a dog.
Google Intro to ML - What is ML?
Classification models can be divided into what two groups?
Binary Classification and Multiclass Classification models.
Google Intro to ML - What is ML?
What is a “binary classification” model?
Binary classification models output a value from a class that contains only two values.
For example, a model that outputs either rain or no-rain
Google Intro to ML - What is ML?
What is a “multiclass classification” model?
Multiclass classification models output a value from a class that contains more than two values.
For example, a model that can output either rain, hail, snow or sleet.
Google Intro to ML - What is ML?
If you wanted to use an ML model to predict energy usage for commercial buildings, what type of model would you use a classification model, or a regression model?
You should use a regression model because “energy consumption” is likely a numeric value such as 24KWH rather than a binary value (e.g. “very little energy consumption” or “a lot of energy consumption”), or even a multiclass value (e.g. “light consumption”, “moderate consumption”, “heavy consumption”).
Google Intro to ML - What is ML?
What is “Unsupervised Learning”?
Unsupervised learning is when a model is trained on data that does not contain any correct answers (a.k.a labels). An unsupervised learning model’s goal is to identify meaningful patterns among the data.
Google Intro to ML - What is ML?
What is a common unsupervised learning technique?
Clustering
Google Intro to ML - What is ML?
What is “clustering” in unsupervised learning?
Clustering is when an unsupervised learning model is used to find data points that demarcate natural groups.
For example, an unsupervised learning model might cluster a weather dataset based on temperature, revealing segmentations that define the seasons.
Google Intro to ML - What is ML?
What is the distinction between how supervised and unsupervised models are trained?
Supervised models are trained on data that includes the correct answers (a.k.a. labels) whereas unsupervised models are trained on data without the correct answers (a.k.a un-labeled).
