overfit
verbEtymology From over- + fit.
1
Mathematics, Sciences, Statistics To use a statistical model that has too many parameters relative to the size of the sample, leading to a good fit with the sample data but a poor fit with new data.
- In all modeling techniques, and neural networks in particular, care must be taken not to overtrain or overfit the model. If possible, models should be interpreted in terms of their mechanistic meaning.2004, Mark T.D. Cronin, Predicting Chemical Toxicity and Fate, CRC Press, →ISBN, page 169:
- If your strategy is to always fit and use the global model, you will probably overfit the model (i.e., include unnecessary variables).2007, Kenneth P. Burnham, David R. Anderson, Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, Springer Science & Business Media, →ISBN, page 250:
- Related terms
- overfit stream