AIExplainer
Machine Learning Intermediate 2 min read

What is ground truth?

The actual, true state of something, used as a reference point to evaluate the accuracy of predictions or models

Ground truth refers to the real, observed data or facts about a situation, which serves as a baseline to measure the performance of a model or algorithm. It is the ultimate reality against which predictions or estimates are compared

Think of ground truth like a measuring tape. Just as a measuring tape gives you the actual length of an object, ground truth provides the actual, true value of something, allowing you to compare it to the predicted value

In self-driving cars, the ground truth would be the actual location and movement of the car, as measured by GPS and sensors, which is used to evaluate the accuracy of the car's navigation system

Ground truth is used in machine learning and data science to train, test, and evaluate models. It helps to identify biases, errors, and areas for improvement in the model

Some people think that ground truth is always easily available or perfectly accurate, but in reality, it can be difficult to obtain or may contain some degree of error or uncertainty

The term 'ground truth' originated in the field of remote sensing, where it referred to the actual conditions on the ground, as opposed to the data collected by satellites or other remote sensing technologies

actual value true state reference point baseline

Three products for different needs — explore what’s relevant to you.