We have informally argued that the AdaBoost algorithm uses the weighting

mechanism to force the weak learner to focus on the problematic examples

in the next iteration. In this question we will nd some rigorous justi cation

for this argument. Show that the error of ht w$.$r$.$t$.$ the distribution D$($t$+1)$ is exactly $12.$ That

is$,$ show that for every t $[$T$]$

m

i$=1$

D$($t$+1)$

i

$1$

$[$yi$=$ht$($xi$)]=12$