overfit

American  

[ohv-er-fit] / ˈoʊv ərˌfɪt /

verb (used with object)

overfitted, overfitting

1. Artificial Intelligence. to model (training data) such that the resulting system captures noise, incidental patterns, or idiosyncratic features of the training set, thereby reducing its ability to generalize accurately.

---

Other Word Forms

• overfitment noun
• overfitted adjective

Example Sentences

A possible answer comes from artificial neural networks, which face a ubiquitous issue during learning: they can overfit to a particular data set.

From Salon

Richards suggests that the brain’s ability to forget might prevent an effect known as overfit: in the field of artificial intelligence, this is defined as when a mathematical model is so good at matching the data it has been programmed with that it is unable to predict which data might come next.

From Nature

"Not only is the phenomenon well understood; its fundamental behaviors are critical to being an effective data scientist. All data scientists acknowledge that algorithms will overfit to biases without controls. There’s an entire subfield dedicated to studying ways to avoid this."

From Fox News

In people with autism, however, the precision may have a tendency to jump to a high level or get stuck there — for whatever reason, the brain tends to overfit.

From Science Magazine

Babak Hojdat, co-founder of Sentient Technologies, an AI startup with a hedge-fund arm, says that, left to their own devices, machine-learning techniques are prone to “overfit”, ie, to finding peculiar patterns in the specific data they are trained on that do not hold up in the wider world.

From Economist