AIExplainer

What is cross-entropy?

A measure of difference between predicted and actual outcomes

Cross-entropy is a mathematical concept used in machine learning to evaluate the performance of a model. It measures how well the model's predictions match the actual outcomes. The goal is to minimize cross-entropy, which means the model is making accurate predictions.

Imagine you're trying to guess the location of a hidden treasure. Cross-entropy is like a score that measures how far off your guesses are from the actual location. The lower the score, the closer you are to finding the treasure.

In image recognition, cross-entropy can be used to train a model to distinguish between different types of animals. The model makes predictions, and cross-entropy measures how well those predictions match the actual labels.

Cross-entropy is commonly used as a loss function in machine learning algorithms, particularly in classification problems. It helps the model learn from its mistakes and improve its predictions over time.

Some people think cross-entropy is a measure of accuracy, but it's actually a measure of the difference between predicted and actual probabilities.

The concept of cross-entropy was first introduced in the 1940s by Claude Shannon, an American mathematician and electrical engineer, as part of his work on information theory.

log loss Kullback-Leibler divergence

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