Error-Bias DecompositionIt refers to the process of decomposing the integrated generalization error, which can be expressed as follows:

$latex {E= \overline {E}- \overline {A}}$

In this formula, E on the left represents the generalization error after integration, and $latex {\overline {E}}$ is the average generalization error of the individual learners, $latex {\overline {A}}$ It represents the ensemble divergence of individual learners. From this formula, we can conclude that the higher the accuracy and diversity of individual learners, the better the ensemble effect.

References

【1】Let’s read Xiguashu together: Chapter 8 Ensemble Learning