Abstract:This paper establishes a lower bound on the number of states necessary in the worst case to simulate an $n$-state two-way nondeterministic finite automaton (2NFA) by a one-way unambiguous finite automaton (UFA). It is proved that for every $n$, there is a language recognized by an $n$-state 2NFA that requires a UFA with at least $\sum_{k=1}^{n} (k - 1)! \cdot k! \cdot \mathrm{stirling2}(n, k) \cdot \mathrm{stirling2}(n+1, k)$ = $\Omega \big( n^{2n+2} / e^{2n} \big)$ states, where $\mathrm{stirling2}(n, k)$ denotes Stirling's numbers of the second kind. This result is proved by estimating the rank of a certain matrix, which is constructed for the universal language for $n$-state 2NFAs, and describes every possible behaviour of these automata during their computation.

What problem does this paper attempt to address?

### The problems the paper attempts to solve The paper aims to determine the minimum number of states required to simulate a two - way nondeterministic finite automaton (2NFA) with $n$ states as a one - way unambiguous finite automaton (UFA). Specifically, the paper proves that for every $n$, there is a language recognized by a 2NFA with $n$ states, which requires at least the following number of states to be simulated by a UFA: \[ \sum_{k = 1}^{n}(k - 1)!k!\left\{ {n\atop k} \right\}\left\{ {n + 1\atop k} \right\}=\Omega\left(\frac{n^{2n+2}}{e^{2n}}\right) \] where $\left\{ {n\atop k} \right\}$ represents the Stirling numbers of the second kind. ### Key point analysis 1. **Problem background**: - There has always been a problem in the theory of finite automata: how many states does one type of finite automaton need to simulate another type of $n$-state finite automaton. - In particular, the question of whether a two - way nondeterministic finite automaton (2NFA) can be simulated by a deterministic two - way finite automaton (2DFA) with a polynomial number of states is related to the L vs. NL and L/poly vs. NL problems in complexity theory. 2. **Existing achievements**: - Existing research includes converting two - way automata into one - way automata, the complexity of complementary languages, etc. - For specific cases (such as unary alphabets), in - depth research and improvements have also been carried out. 3. **New contributions**: - This paper establishes a lower bound by estimating the rank of a certain matrix, which describes all possible behaviors. - The new lower bound proposed is: \[ \sum_{k = 1}^{n}(k - 1)!k!\left\{ {n\atop k} \right\}\left\{ {n + 1\atop k} \right\} \] - This result is closer to the upper bound $2^{n^{2}-O(n)}$ than the previously known lower bounds and shows the difference between the two transition functions. ### Conclusion Through strict mathematical derivations and matrix analysis, the paper significantly improves the lower bound of the state complexity of the conversion from 2NFA to UFA, thus filling the knowledge gap in this field and providing a new perspective for further research.

A lower bound on the state complexity of transforming two-way nondeterministic finite automata to unambiguous finite automata

Blow-up in Non-Deterministic Automata

Unambiguity and Fewness for Nonuniform Families of Polynomial-Size Nondeterministic Finite Automata

Descriptional Complexity of Finite Automata -- Selected Highlights

Random Deterministic Automata With One Added Transition

Unambiguous and Co-Nondeterministic Computations of Finite Automata and Pushdown Automata Families and the Effects of Multiple Counters

Complexity of Unary Exclusive Nondeterministic Finite Automata

On the Emptiness Problem for Two-Way NFA with One Reversal-Bounded Counter

Descriptional Complexity of Non-Unary Self-Verifying Symmetric Difference Automata

Converting Finite Width AFAs to Nondeterministic and Universal Finite Automata

Rice-like complexity lower bounds for Boolean and uniform automata networks

i dfa: A novel deterministic finite automaton without state explosion

Exact descriptional complexity of determinization of input-driven pushdown automata

State complexity of halting, returning and reversible graph-walking automata

TFA : A Tunable Finite Automaton for Regular Expression Matching

Lifting query complexity to time-space complexity for two-way finite automata

Language Acceptors with a Pushdown: Characterizations and Complexity

An Analysis of On-the-fly Determinization of Finite-state Automata

TFA: A Tunable Finite Automaton for Pattern Matching in Network Intrusion Detection Systems.

On Two-Way Nondeterministic Finite Automata with One Reversal-Bounded Counter

Deriving Nonuniform Lower Bounds from Uniform Nondeterministic Lower Bounds