All terms
Foundations
Bias-Variance Tradeoff
The balance between a model being too simple (bias) and too sensitive to its data (variance).
Definition
The bias-variance tradeoff is a core machine-learning idea describing two ways models fail. High bias means the model is too simple and misses real patterns (underfitting); high variance means it is too tuned to the training data's quirks and fails on new data (overfitting). Good models balance the two.