The only explanations I've seen of the bias/variance tradeoff rely on rewriting the squared error of an estimator as the sum of bias and variance terms. How does the bias/variance tradeoff work if the loss function is not squared error? Thanks in advance!
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.
Here's how it works:
- Anybody can ask a question
- Anybody can answer
- The best answers are voted up and rise to the top
MSE=bias$^2$ +variance So the tradeoff is obvious for MSE. If your loss function is some other function of bias and variance then there will be a variance-bias tradeoff for it too. Otherwise there is none for varying the expected loss.