Explain like I'm five
Imagine you're studying for a test with a stack of flashcards. One epoch is like going through every single card once. After you finish the whole stack, you've completed one epoch, and you might do it again to learn better.
Artificial Intelligence Difficulty 50/100
Epoch
One full lap around the data.
An epoch is one complete pass of the entire training dataset through a machine learning model.
Why it matters
Epochs control how many times the model sees the data, which directly affects how well it learns. You encounter it whenever you train a neural network, like for image recognition or language translation.
Common misconception
People often think more epochs always mean better learning, but too many can cause overfitting, where the model memorizes data instead of generalizing. It's not about repetition alone, but finding the right balance.
Formal definition
In machine learning, an epoch is a hyperparameter that defines the number of times the learning algorithm processes the entire training dataset. One epoch means each sample in the dataset has been used to update the model's parameters exactly once. Multiple epochs are typically required for convergence.