Glossary term
Glossary term
Training and Fine-Tuning
Overloaded term that can mean either of the following:
Reducing the amount of information in a feature in order to train a model more efficiently. For example, before training an image recognition model, downsampling high-resolution images to a lower-resolution format.
Training on a disproportionately low percentage of over-represented class examples in order to improve model training on under-represented classes. For example, in a class-imbalanced dataset, models tend to learn a lot about the majority class and not enough about the minority class. Downsampling helps balance the amount of training on the majority and minority classes.
See Datasets: Imbalanced datasets in Machine Learning Crash Course for more information.
Created for this library
A fraud team downsamples the dominant non-fraud class during training so the gradient signal is not overwhelmed by majority examples.
A search ranking team downsamples non-clicked impressions in its training data to make the dataset smaller without losing signal on positives.
A medical imaging team downsamples normal cases when training a rare-disease detector to keep the class balance manageable.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License