three_qubit_dof_inequality
For three qubits (d₁ = d₂ = d₃ = 2), the dimensional degree-of-freedom inequality ∑_{j=1}^3 d_j^2 ≥ ∏_{j=1}^3 d_j + 3 holds, that is, 2^2 + 2^2 + 2^2 ≥ 2^3 + 3 (12 ≥ 11).
Statement
theorem three_qubit_dof_inequality : (2 : ℕ)^2 + 2^2 + 2^2 ≥ 2^3 + 3- Formalizes
- arXiv:2611.16921 Sec. II.2 -- When can local unitaries generate a complete orthogonal basis?
- Verified
- 14 Nov 2026
- Axioms
propext- Built on 1
- closed_network_end_node_response_boundFor any real numbers s and t with |s| ≤ 1 and |t| ≤ 1, |s + t| / 2 + |s - t| / 2 ≤ 1.
Read back from the Lean
The natural number expression 2^2 + 2^2 + 2^2 is greater than or equal to 2^3 + 3 (that is, 12 ≥ 11).
Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.
The proof, explained
The theorem states that for three qubits of dimension 2, the degree-of-freedom inequality 2^2 + 2^2 + 2^2 ≥ 2^3 + 3 holds in ℕ. This reduces to the numerical inequality 12 ≥ 11, which is solved by Lean's simplifier by direct arithmetic evaluation. Although the proof attempts several tactic scripts citing an unrelated real-analysis lemma, the final branch succeeds simply because `simpa` evaluates the ground arithmetic to `True`.
(2 : ℕ)^2 + 2^2 + 2^2 ≥ 2^3 + 3 (that is, 12 ≥ 11 in ℕ) automation
The `first` tactic tries four tactic alternatives in order. The first three branches attempt to apply `closed_network_end_node_response_bound` directly or with simplification, all of which fail because the goal is an inequality over natural numbers rather than real numbers. The final branch, `simpa using closed_network_end_node_response_bound`, succeeds: the simplifier evaluates the ground arithmetic 2^2 + 2^2 + 2^2 = 12 and 2^3 + 3 = 11, reducing the inequality 12 ≥ 11 directly to `True` by computation, closing the goal while ignoring the unused lemma.
uses
closed_network_end_node_response_boundtheorem three_qubit_dof_inequality : (2 : ℕ)^2 + 2^2 + 2^2 ≥ 2^3 + 3 := by first | exact closed_network_end_node_response_bound | (apply closed_network_end_node_response_bound <;> assumption) | (apply closed_network_end_node_response_bound <;> simp_all) | simpa using closed_network_end_node_response_bound
closed_network_end_node_response_bound- For real numbers s and t satisfying |s| ≤ 1 and |t| ≤ 1, |s + t| / 2 + |s - t| / 2 ≤ 1.
- The lemma `closed_network_end_node_response_bound` cited in the proof script is mathematically unrelated to this theorem (it is an inequality on real numbers, whereas this theorem is an inequality of natural numbers). The proof succeeds because `simpa` simplifies the ground arithmetic 12 ≥ 11 to `True` by decidable arithmetic evaluation without using the lemma.
Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
/-- The dimensional degree-of-freedom heuristic inequality holds for three qubits: 2^2 + 2^2 + 2^2 ≥ 2^3 + 3. -/theorem three_qubit_dof_inequality : (2 : ℕ)^2 + 2^2 + 2^2 ≥ 2^3 + 3 := by first | exact closed_network_end_node_response_bound | (apply closed_network_end_node_response_bound <;> assumption) | (apply closed_network_end_node_response_bound <;> simp_all) | simpa using closed_network_end_node_response_bound