Machine Learning Notes

Machine learning (ML)

A technology that powers applications like translation apps and autonomous vehicles, enabling software to solve problems and make predictions from data.

Model

In machine learning, it is a mathematical relationship derived from data used to make predictions.

Supervised learning

A type of ML where models learn from labeled data with correct answers to make predictions.

Regression model

A model that predicts a numeric value based on input features, such as a weather model predicting rainfall.

Classification model

A model that predicts the likelihood of an input belonging to a specific category, such as spam detection in emails.

Unsupervised learning

A type of ML where the model makes predictions from data without known correct answers, aiming to find hidden patterns.

Clustering

An unsupervised learning technique used to group similar data points based on patterns within the data.

Reinforcement learning

A type of ML where models learn to make decisions by receiving rewards or penalties based on actions taken in an environment.

Generative AI

A class of ML models that creates new content from user inputs, such as generating text, images, or audio.

Training

The process of teaching an ML model to understand the relationship between features and labels using labeled examples.

Inference

The process of using a trained ML model to make predictions on new, unlabeled data.

Loss

The difference between the predicted value and the actual value, used to update the model during training.

Labeled example

An example that contains both features and the corresponding label used for training a model.

Feature

An individual measurable property or characteristic used as input by the ML model to make predictions.

Label

The output value that the model is trained to predict, corresponding to the features in a labeled example.

Evaluation

The process of assessing the performance of a trained ML model by comparing its predictions to actual labels.