MHB Equivalence Between NFA and DFA: Q, Q` & $\Sigma$

  • Thread starter Thread starter mathmari
  • Start date Start date
  • Tags Tags
    Dfa Equivalence
AI Thread Summary
The discussion focuses on the equivalence between a Non-deterministic Finite Automaton (NFA) and a Deterministic Finite Automaton (DFA). It establishes that for states Q_I and Q_j in the DFA, the condition for Q_I transitioning to Q_j with input 'a' is that Q_j equals the union of states reachable from Q_I after processing 'a' and any number of ε transitions. Specifically, Q_j can be expressed as the set of states that can be reached from any state in Q_I through a sequence of transitions involving 'a' followed by ε transitions. This relationship is critical for understanding how NFAs and DFAs can be converted into one another. The discussion emphasizes the importance of ε transitions in determining state reachability in the context of automata theory.
mathmari
Gold Member
MHB
Messages
4,984
Reaction score
7
Heloo! :o

I am looking at the equivalence between a NFA and a DFA.

NFA: Q={q1,q2}
DFA: Q`=P(Q}

When $a\in \Sigma, Q_I, Q_j \in Q`$ which is sufficient and necessary condition so that $ Q_I \overset{a}{\rightarrow}Q_j$?
 
Last edited by a moderator:
Physics news on Phys.org
mathmari said:
When $a\in \Sigma, Q_I, Q_j \in Q`$ which is sufficient and necessary condition so that $ Q_I \overset{a}{\rightarrow}Q_j$?
This is described in the books we talked about. Namely, if $R\in Q'$, i.e., $R\subseteq Q$, let
\[
E(R)=\{q\in Q\mid q\text{ can be reached from }R\text{ by 0 or more }\varepsilon\text{ arrows}\}.
\]
Then $Q_i \overset{a}{\rightarrow}Q_j$ iff $$Q_j=\bigcup_{q\in Q_i}E(\delta(q,a))$$. In other words,
\[
Q_j=\{s\in Q\mid q\overset{a}{\to}r\overset{\varepsilon}{\to}\!\!^*s\text{ for some }q\in Q_i\}.
\]
Here $r\overset{\varepsilon}{\to}\!\!^*s$ means that $s$ can be reached from r by 0 or more $\varepsilon$ arrows.
 
Evgeny.Makarov said:
This is described in the books we talked about. Namely, if $R\in Q'$, i.e., $R\subseteq Q$, let
\[
E(R)=\{q\in Q\mid q\text{ can be reached from }R\text{ by 0 or more }\varepsilon\text{ arrows}\}.
\]
Then $Q_i \overset{a}{\rightarrow}Q_j$ iff $$Q_j=\bigcup_{q\in Q_i}E(\delta(q,a))$$. In other words,
\[
Q_j=\{s\in Q\mid q\overset{a}{\to}r\overset{\varepsilon}{\to}\!\!^*s\text{ for some }q\in Q_i\}.
\]
Here $r\overset{\varepsilon}{\to}\!\!^*s$ means that $s$ can be reached from r by 0 or more $\varepsilon$ arrows.

Ok... Thanks a lot! (Smile)
 
Back
Top