AI Dictionary › Fondamenti AI

Epoch

Epoca

An epoch is one complete pass of the training algorithm over the entire available dataset. If the model has seen every example in the dataset exactly once, it has finished one epoch. It's like studying a whole book cover to cover: reading it once rarely suffices, and to truly absorb it you reread it several times. Likewise a model passes through the dataset for many epochs, refining parameters each round. Each epoch consists of many smaller steps, one per mini-batch, where prediction, loss computation, and gradient-descent updates occur.

Definition

The number of epochs is a delicate choice: too few and the model stays underprepared, too many and it risks memorizing the data instead of generalizing, falling into overfitting. Tracking error on unseen data across epochs helps stop training at the right moment.

Related terms

• Gradient Descent
• Loss Function
• Overfitting
• Training Data
• Machine Learning

More in Fondamenti AI

• LLM (Large Language Model)
• Token
• Context Window
• Temperature
• AI Hallucination
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Embedding

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.