How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Epsilon-NFAs, finite epsilon closure, and extended word transition
Definition
An epsilon-NFA is a tuple with finite , , , a finite alphabet , and a total map Here the disjoint union tags letters and the empty-input label separately; on an edge consumes no letter. Missing arrows in a diagram mean empty transition sets. Words and concatenation have the convention of Computation alphabets, words, the empty word, and .
For , put and define Write and . A subset is epsilon-closed when for every . The construction below shows that is the least epsilon-closed superset of .
The extended word transition is the unique function satisfying Abbreviate . A finite path of length is a sequence of states together with labels such that . Its spelling is the word obtained by erasing the epsilon labels. Length-zero paths are allowed. The set consists exactly of endpoints of such paths starting in with spelling .
The closure is extensive, monotone, idempotent, preserves every set-indexed union of subsets of , and satisfies . A word is accepted when , and
Facts & Assumptions
Given: The tuple above, , and . All constructions use ZF; no choice axiom is assumed.
Power sets contain precisely the subsets of their argument, and indexed unions have the existential membership rule (The power set , , and for ).
A subset of a finite set is finite, its cardinality is no larger, and equality occurs exactly for the whole set (A subset of a finite set is finite, with , and equality holds if and only if ). Cardinalities are natural numbers (The cardinality of a finite set).
A self-map of a set, with a specified initial element, has a unique natural-number iterate sequence (The recursion theorem).
A property true at zero and preserved by successor holds at every natural number (The principle of mathematical induction).
Words are functions , including the unique empty word, and is a set (Computation alphabets, words, the empty word, and ).
A language over is a subset of (Languages over an alphabet).
Cartesian products are sets of the specified ordered pairs (The Cartesian product ); Separation forms the subset of an ambient set satisfying a formula with fixed parameters (The Axiom Schema of Separation: for each formula , ).
Verification
For , the union of the sets with is a subset of . Thus is a self-map of . Applying natural-number recursion with initial element gives the unique sequence , with .
If , applying repeatedly gives for every natural : the case is equality and the successor follows from . If , therefore all inclusions from through are strict. Each raises natural cardinality by at least one, by the finite-subset equality criterion. Starting at , induction gives , contradicting and . Hence and all subsequent stages equal .
The equality says precisely that is epsilon-closed, and . If and is epsilon-closed, then ; from , every successor of a state in is in , so . Induction gives . This proves leastness and extensivity.
For every , is exactly the set of endpoints of epsilon paths from with at most edges. At both sets are . An endpoint in either already lies in , or follows one epsilon edge from a state in ; retaining or appending that edge gives a path of length at most . Conversely a path of length at most either has length zero, or deleting its last edge leaves a path of length at most , whose endpoint belongs to ; its last endpoint lies in . These implications give the successor equivalence and induction proves the claim. Stabilization then implies that is exactly the endpoints of all finite epsilon paths: paths of length at most give , while longer paths give a stage equal to .
If , the closed set contains , so leastness gives . Because is already closed, leastness gives and extensivity gives the reverse inclusion. For any family , a path starts in exactly when its starting state belongs to some . Applying the endpoint equivalence gives in both directions. For neither side has a path starting point, so .
Fix . On define if , and if . In the first clause exists; in both clauses the value lies in the product. Recursion starting at gives a unique sequence. For its value is , as follows by induction from its first clause. In particular and for . Define . The graph is the subset of consisting of the triples satisfying this finite-sequence prescription, so Separation gives a set function; existence and uniqueness of the sequence give exactly one output for each input. Here the ambient products are sets; the finite-sequence condition is a set-theoretic formula with parameters .
If two words agree at positions below , their sequences for the same agree through stage : the initial sets agree, and equality at stage gives equal unions with the same letter, hence equal next closures. Induction proves this prefix compatibility. Taking the words and proves the displayed successor equation, while proves the empty-word equation. Any other function satisfying those equations agrees at the empty word and, if it agrees at , agrees at by substitution. Every nonempty word has this last-letter decomposition, so induction on length proves uniqueness on all words.
For spelling , the endpoint assertion is the epsilon-path assertion. Suppose the assertion holds for and consider . A path spelling has a last non-epsilon edge: its label is , its preceding path spells , and its following path has only epsilon labels. The preceding endpoint is in , so the last edge lands in and its epsilon suffix lands in the closure of that union. Conversely a state in this closure has an epsilon path from some with . The induction hypothesis supplies a path spelling to ; concatenating this path, the edge to , and that epsilon path gives spelling . Only witnesses for this one membership are used. Induction proves both directions for all words.
Consequently the acceptance predicate holds exactly when a finite path from spelling ends in . Its defining set is a subset of , hence a language. Empty has no path starts, so for every word; empty accepts no word. If the only word is , already treated. A one-state and epsilon self-loops satisfy the same stabilization argument; is excluded by . Thus the definitions and all asserted properties are well-defined in the boundary cases as well.
Source conventions
Gallier, §§3.3–3.5, Definitions 3.5–3.7, pp. 28–33 supplies the NFA convention and saturation route; Aho, Lecture 3 §§1–2 gives the corresponding set closure. The existence, uniqueness, closure laws, and path equivalence used here are derived above. The tagged epsilon label avoids any set-theoretic collision with letters.
Depends on
- Computation alphabets, words, the empty word, and $\Sigma^*$
- The power set $\mathcal{P}(x) = \{\, z : z \subseteq x \,\}$
- $\bigcup_{i \in I} A_i := \bigcup \{A_i : i \in I\}$, and $\bigcap_{i \in I} A_i := \bigcap \{A_i : i \in I\}$ for $I \neq \varnothing$
- The cardinality $\lvert A\rvert$ of a finite set
- A subset of a finite set is finite, with $\lvert B\rvert \le \lvert A\rvert$, and equality holds if and only if $B = A$
- The recursion theorem
- The principle of mathematical induction
- Languages over an alphabet
- The Cartesian product $A \times B := \{\, z \in \mathcal{P}(\mathcal{P}(A \cup B)) : \exists a \in A\ \exists b \in B\ z = (a,b) \,\}$
- The Axiom Schema of Separation: for each formula $\varphi$, $\forall \bar p\,\forall x\,\exists y\,\forall z\,(z \in y \leftrightarrow (z \in x \wedge \varphi(z,\bar p)))$
Used by
- Dropping epsilon closure changes the accepted language Counterexample
- A complete determinisation table for a three-state epsilon-NFA Example
- Epsilon closure and word transitions on three states Example
- A DFA embeds as an epsilon-free NFA Proposition
- Subset construction and its exact reachability invariant Theorem
Dependency tree · two levels
38 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.
Sources
- Jean Gallier, Introduction to the Theory of Computation (2018), §§3.3–3.5 (standard reference, not scraped)
- Alfred Aho, COMS W3261 Lecture 3, §§1–2 (standard reference, not scraped)