← Glossary · Training and Tuning

training

Concept

Fact-checked May 20, 2026

Also called: model training, AI training

Training is the process where an AI model learns from data to identify patterns and make predictions or decisions.

In AI, training is like teaching a student with many examples. You feed a model a huge amount of data, and it gradually adjusts its internal settings to recognize relationships, patterns, and features within that data. This process involves complex mathematical calculations and algorithms that minimize errors over time, allowing the model to get better and more accurate at its intended task.

Think of training a model to recognize pictures of cats. You would show it thousands of images, some with cats and some without. During training, the model tries to guess if a cat is present, and if it's wrong, it learns from its mistake and updates its internal parameters. Eventually, after seeing enough examples and correcting itself, the model becomes proficient at identifying cats in new, unseen images.

Related terms

CutMixCutMix is a data augmentation technique that helps train more robust image classification models by blending parts of two different images.fine-tuningFine-tuning is a process where a pre-trained AI model is further trained on a smaller, specific dataset to adapt its knowledge to a particular new task or domain.Multi-Task LearningMulti-Task Learning (MTL) is a machine learning technique where a single model learns to perform several related tasks at the same time.RLHFRLHF is a way to train AI models using human preferences, helping the AI learn to produce outputs that are more helpful and safe.Stochastic Gradient DescentSGD is a popular optimization algorithm used to train machine learning models, especially neural networks.

Learn AI in 5 minutes a day.

Daily Deck explains terms like training as part of a free seven-card daily brief. No jargon. No fluff.

Start free