Theorem 19.4: for integer c ≥ 2 and any learning rule, some distribution on [0,1]^d × {0,1} with c-Lipschitz η and Bayes error 0 forces expected error ≥ 1/4 when 2m ≤ (c+1)^d

Initializing...

Theorem 19.4: for integer c ≥ 2 and any learning rule, some distribution on [0,1]^d × {0,1} with c-Lipschitz η and Bayes error 0 forces expected error ≥ 1/4 when 2m ≤ (c+1)^d · Prove2Me