What are Bias Errors?
Bias are the simplifying assumptions created by a model to produce the target function more comfortable to learn.
Bias Variance Tradeoff
- Low Bias - Suggests fewer algorithms about the form of the target function. For example, Decision Trees, k-Nearest Neighbors, and Support Vector Machines.
- High Bias - Suggests more algorithms about the form of the target function. For example, Linear Regression, Linear Discriminant Analysis, and Logistic Regression.