We want to learn an unknown monotone conjunction c with n variables. For that, we draw i.i.d. m assignments that are correctly labelled as 0 (false) or 1 (true). We want to use the VC dimension to calculate the sample size for PAC-learnability. (a) Prove that the VC dimension of the hypothesis class formed by monotone conjunctions with n variables is at least n. You need to prove that there is a set of n assignments that can be shattered with monotone conjunctions,

icon
Related questions
Question
100%
We want to learn an unknown monotone conjunction c with n variables. For that, we draw i.i.d. m assignments that are
correctly labelled as 0 (false) or 1 (true). We want to use the VC dimension to calculate the sample size for PAC-learnability.
(a) Prove that the VC dimension of the hypothesis class formed by monotone conjunctions with n variables is at least n.
You need to prove that there is a set of n assignments that can be shattered with monotone conjunctions,
Transcribed Image Text:We want to learn an unknown monotone conjunction c with n variables. For that, we draw i.i.d. m assignments that are correctly labelled as 0 (false) or 1 (true). We want to use the VC dimension to calculate the sample size for PAC-learnability. (a) Prove that the VC dimension of the hypothesis class formed by monotone conjunctions with n variables is at least n. You need to prove that there is a set of n assignments that can be shattered with monotone conjunctions,
Expert Solution
steps

Step by step

Solved in 2 steps

Blurred answer