Overfitting — Overfitting happens when a model learns its training data too closely, including its noise and quirks. It performs well on training examples but poorly on new data. Overfitting is like memorizing answers instead of understanding a subject. More data and careful training reduce it.

Added October 21, 2026