Variance & Bias

Variance & Bias

• Bias - the difference between the average prediction of model and the correct value(center of Target)
• Variance - variability of model prediction(예측값들의 분산된 정도)

Underfitting model?

• usually have high bias and low wariance
• happens when have very less data or try to build a linear model with a nonliner data

Overfitting model?

• usually have low bias and high variance
• happens when our model captures the noise along with the underlying pattern in data(train model a lot over noisy dataset)

Reference

https://medium.com/@toprak.mhmt/the-bias-variance-tradeoff-d9320282ac04