Short Definition
The bias-variance tradeoff describes the balance between models that are too simple to capture structure and models that are too sensitive to training data noise.
Intuition
High-bias models miss the pattern. High-variance models chase noise. Good generalization usually needs a balance.
Technical Definition
The tradeoff decomposes expected prediction error into bias, variance, irreducible noise, and interactions depending on the setting.
Example
A linear model may have high bias on nonlinear data, while a very deep tree may have high variance.
Common Misunderstandings
Bias here is statistical error, not social bias.
Modern deep learning complicates the simple textbook U-shaped curve, but the intuition remains useful.