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.
Logarithmic Space, NL, and Reachability: Examples and Counterexamples
1 · Prerequisites
- Construction of the Natural Numbers
- Construction of the Real Numbers via Cauchy Sequences
- Construction of the Real Numbers via Dedekind Cuts
- Countability and Uncountability
- Eulerian and Hamiltonian Graphs
- Finite Counting, Factorials and Binomial Coefficients
- Formal Languages, Encodings, and Decision Problems
- Foundations of the Real Numbers for Analysis
- Graphs, Walks and Connectivity
- Linear Recurrences and Rational Generating Functions
- Logarithmic Space, NL, and Reachability
- Relations, Functions, and Quotients
- Resource Bounds and Machine Invariance
- Sequences and Limits
- The ZFC Axioms and the Basic Set Constructions
- Time and Space Hierarchy Theorems
- Turing Machines, Configurations, and Computation
2 · Summary
These worked cases expose the virtual-output-tape convention, a concrete tableau-to-circuit compilation, and the existential-branch error in trying to complement nondeterminism by terminal-state swapping.
3 · Logical flowchart
4 · Definitions, theorems and proofs
None yet.
5 · Examples, counterexamples and false statements
Composing two logspace reductions through a virtual tape
Example
Work over . Let . Let output the first and last symbols of when , and set . Both total functions are deterministic logspace transductions. To compute , a composed machine asks for virtual cells and rather than storing the string . Let be the words starting with , let be the nonempty words starting with and ending with , and let . Then reduces to and reduces to .
Facts & Assumptions
Given: an input of length .
A reduction output can be recomputed at a requested position in logarithmic work space (Logspace many-one reduction).
If and , then (Logspace reductions compose).
Verification
The first requested virtual symbol is . Maintain a counter to find and recompute the final virtual symbol, which is when and when . Forward and backward scans emit , while two endpoint scans emit ; both use at most logarithmic work space and halt on every input.
Thus the composed output is for and for , exactly ; it used only counters and a reread of .
The word lies in exactly when starts with , and exactly when starts with and ends with ; the empty input maps to . Thus these are total logspace reductions , and the explicit virtual computation illustrates their composition in [L2].
A small computation tableau compiled to Circuit Value
Example
On inputs promised to have length two, consider a fixed machine that reads the two bits and accepts exactly . A circuit for this computation uses , , and output . For , the values are .
Facts & Assumptions
Given: the fixed machine on two-bit inputs and input .
Every fixed polynomial-time machine has a polynomial-size logspace-uniform circuit family deciding its acceptance (Polynomial-time computations have logspace-uniform circuit families).
Circuit Value is P-complete under logspace many-one reductions (Circuit Value is P-complete).
Verification
On , and , agreeing with the machine's accepting final state.
For every two-bit input, exactly when and . Number the input nodes and the NOT and AND nodes , with predecessor lists and and output node . This fixed circuit can be emitted in constant work space; it is a direct circuit for the promised computation, illustrating the existence assertion in [L1].
The pair is therefore a yes-instance of Circuit Value, illustrating the acceptance-preserving reduction in [L2].
A nondeterministic branch tree where state swapping is not complementing
Statement refuted
State swapping complements a nondeterministic machine's language.
Facts & Assumptions
Given: a fixed input and a nondeterministic machine which, from its start configuration on , has exactly two choices: one moves directly to an accepting state and the other directly to a rejecting state.
A nondeterministic machine accepts when at least one branch accepts (A read-only-input logarithmic-space machine).
The state-swapping inference is the false statement refuted on the A page (NL equals coNL follows by swapping accepting and rejecting states).
Counterexample
The original accepting leaf makes the original machine accept by [L1].
After labels swap, the original rejecting leaf is accepting, so [L1] makes the swapped machine accept as well.
Since is in the original language, it is not in its complement; step 2.1 therefore contradicts the proposed complementing behavior, as asserted in [L2].