文章目录
一、机器学习模型
二、机器学习算法
三、机器学习目标
1、偏差和方差
- 偏差(Bias):how much the average model over all training sets differ from the true model?
- 方差(Variance):how many models estimated from different training sets differ from each other
2、拟合
欠拟合(Under-fitting):model is too "simple" to represent all the relevant class characteristics.
- 高偏差&低方差
- 高训练误差&高测试误差
过拟合(Over-fitting):model is too "complex" and fits irrelevantcharacteristics (noise) in the data.
- 低偏差&高方差
- 低训练误差&高测试误差