0 votes 0 votes how to construct pda for a^i b^j where j!=2i+1, i>=0 ? i have constructed for j=2i+1 but by complementing the states do we get actual n pda of our requirement? whether it is decidable? Theory of Computation pushdown-automata + – Ravi Raaja asked Sep 17, 2015 • retagged Jul 4, 2017 by Arjun Ravi Raaja 2.3k views answer comment Share Follow See all 0 reply Please log in or register to add a comment.
0 votes 0 votes Construct for j>2i+1 and j<2i+1 then take the union of these two pda that will be j!=2i+1. Umang Raman answered Sep 28, 2015 Umang Raman comment Share Follow See all 2 Comments See all 2 2 Comments reply Arjun commented Sep 28, 2015 reply Follow Share Nopes. A complement language must include every string in $\Sigma^*$ not in $L$. Here, "ba" should be in $L'$. 0 votes 0 votes Umang Raman commented Sep 28, 2015 reply Follow Share I guess here we have to design for L not the complement of L and yes off course CFL language is not closed under complement so we cant design for complement of a CFL language it would be CSL. Here L is a CFL so npda exist , yes "ba" should be in L' not in L(coz it is a^ib^j) and we are designing for L i think which means "ab" should exist and we design using union it will exist in npda. (please let me know if i am incorrect.) Ll 0 votes 0 votes Please log in or register to add a comment.
0 votes 0 votes Yes, but works for only deterministic PDAs and has problems and won't work straight away- see below link. http://drona.csa.iisc.ernet.in/~deepakd/atc-2011/DPDA.pdf For NPDA, a complement PDA may not even exist. Arjun answered Sep 28, 2015 Arjun comment Share Follow See all 0 reply Please log in or register to add a comment.