What is Training?
Training is the process of feeding data into a machine learning model so it can learn patterns and adjust its internal parameters to make accurate predictions.
In training, the model processes input examples, generates outputs, and measures errors against known correct labels using a loss function.
Optimization techniques such as gradient descent then update the model's weights to reduce those errors over many iterations called epochs.
The goal is to find parameter values that generalize well, turning raw data into a functional predictive system.
Example
A spam filter is trained on thousands of emails already labeled as spam or not spam, allowing it to learn word patterns and sender traits that indicate unwanted messages.
Why it matters
Training turns static data into intelligent behavior, enabling today's AI applications from recommendation engines to medical diagnosis tools.
Frequently asked questions
Training teaches the model using labeled data, while testing evaluates how well it performs on new, unseen data.
Related terms
A dataset is a structured collection of data points used to train, validate, or test machine learning models.
An epoch is one complete pass of a machine learning model through the entire training dataset during training.
A loss function quantifies how far a model's predictions are from the true values, serving as the objective that training tries to minimize.
Gradient descent is an optimization algorithm that finds the minimum of a function by repeatedly moving in the direction of the steepest downward slope. In machine learning it is used to minimize a model's error by adjusting parameters step by step.
Overfitting happens when a machine learning model learns the training data too closely, including its noise and quirks, so it fails to perform well on new, unseen data.
Batch size is the number of training examples processed together in a single forward and backward pass during model training.