What term describes the learning algorithm's iteration through all samples in the dataset?

What term describes the learning algorithm's iteration through all samples in the dataset?

• A. Model training parameters
• B. Epochs
• C. Hyperparameters
• D. Confusion matrix

Answer: (B)

The learning algorithm's iteration through all samples in the dataset is called an epoch. One epoch means the model has seen every training example once, with parameter updates occurring after each batch or after each example depending on the training method. If you use mini-batches, an epoch consists of multiple updates, one per batch. This term is distinct from the model’s learned parameters, which are the weights being adjusted, from hyperparameters like the learning rate, and from evaluation tools such as a confusion matrix.

The learning algorithm's iteration through all samples in the dataset is called an epoch. One epoch means the model has seen every training example once, with parameter updates occurring after each batch or after each example depending on the training method. If you use mini-batches, an epoch consists of multiple updates, one per batch. This term is distinct from the model’s learned parameters, which are the weights being adjusted, from hyperparameters like the learning rate, and from evaluation tools such as a confusion matrix.