please help me with this question! thank you so mcuh

Bias-variance tmdeo' 4. In lecture we present the biasvariance tradeo' that takes the form E [(yo xer] : Var(f(xo)) + [Bias(f(xo))]2 + We), where the underlying model Y : f(X) + E satisifes: (1) 5 is a zero-mean random noise, and X is nonrandom (all randomness in Y comes from 5); (2) (x0, yo) is a test observation, independent of the training set, and drawn from the same model; (3) f () is the estimate of f obtained on a training set. (a) Which of the term(s) in the bias-variance tradeoff above represent the reducible error? Which term(s) represent the irreducible error? (b) Use the biasvariance tradeoff above to show that the expected test error is always at least as large as the irreducible error