gpt 6 astra ultra attempt on Phys-390

Local Density Approximation for the Canonical Functional

Definitions and mathematical statement

Shared notation.

$\mathbb N=\{1,2,\ldots\}$; the standard number fields have their usual meanings. All additional conventions are stated locally.

For a prescribed particle density $\rho$, the canonical functional is a fixed-particle-number constrained search, schematically $F(\rho)=\inf_{\Gamma\mapsto\rho}\operatorname{Tr}[(T+W)\Gamma]$, over admissible fermionic states. The local-density approximation replaces a slowly varying inhomogeneous energy by the integral of a uniform-gas energy evaluated at the local density.

Catalogue wording retained for traceability.

Can the local-density approximation be rigorously justified for the canonical density functional $F(\rho)$?

Scope and source qualification.

The canonical constraint is essential: a grand-canonical result does not automatically settle this question. Scaling, particle number and interaction are those of the report.

Short English statement

Justify the local-density approximation for the fixed-particle-number density functional.

Research attempt

Model, date and investigated scope.

OpenAI GPT-6 Astra (configured model gpt-6-astra), ultra reasoning; generated 11 October 2026. Computed a fermionic example where number projection changes density and an exact density repair changes pair energy.

Sources and literature check.

Read the full supplied definition, catalogue wording, source qualifications and listed references for 30005744. No fresh status or novelty search was performed. The argument below proves its auxiliary statements directly and does not import an unverified external theorem.

Testing projection from fluctuating particle number.

A proposed route to the canonical functional is to build a grand-canonical local-density trial state and project to the desired total number. Even for two disjoint one-mode cells this changes the local density. Let independent occupations $X_1,X_2\in\{0,1\}$ have probabilities $p_1=1/4$, $p_2=3/4$. Their expected total is one, but \[ \mathbb P(X_1=1,X_2=0)=1/16,\quad \mathbb P(X_1=0,X_2=1)=9/16. \] Conditioning on $X_1+X_2=1$ gives local means $1/10$ and $9/10$, instead of $1/4$ and $3/4$. These probabilities define legitimate diagonal fermionic density matrices, so the obstruction is present without coherences or interactions.

A repair in the finite example.

Correlating the original cells instead, use the canonical mixture \[ \Gamma=\tfrac14|10\rangle\langle10|+\tfrac34|01\rangle\langle01|. \] It has exactly one particle and the prescribed local means. For a diagonal one-body energy $h_1X_1+h_2X_2$, its expectation equals the independent state's energy. For an interaction $JX_1X_2$, however, the expectations are respectively zero and $3J/16$. Thus exact density repair is possible here but changes correlations and interaction energy in a controlled, explicitly nonzero way.

Remaining canonical estimate.

The continuum constrained search must preserve the entire density, not just its integral. The first missing step is a number-correlation or density-repair construction with kinetic and interaction costs negligible in the source local-density scaling. This two-cell calculation supplies neither that uniform estimate nor an asymptotic local-density theorem, and no grand-canonical result is substituted for the canonical target.

Verification and adversarial audit.

Checked all conditional probabilities, admissibility of diagonal fermion states, exact total number, density and interaction expectation before and after repair. Model self-audit only; no independent human mathematical verification. Applied contribution prompt A before drafting and B afterwards. Checked all conditional probabilities, admissibility of diagonal fermion states, exact total number, density and interaction expectation before and after repair. First remaining gap: Construct a continuum fixed-number density-preserving trial state whose additional kinetic and interaction costs are negligible in the source scaling. The inherited source status is preserved; original-target completion remains zero.

Coordinator checked the conditioned occupation probabilities and the explicitly changed interaction energy after canonical density repair. This additional check was model-assisted, not independent human verification.

Tools, process and contributions.

Python 3 and local file tools used to read definitions and serialize manuscripts. Calculations are symbolic reasoning; no numerical experiment or theorem prover. Contribution prompt A was followed by prompt B. GPT-6 Astra with ultra reasoning drafted the research and performed its stated self-audit; any additional coordinator contribution is identified in the verification and audit. A GPT-6-based Codex coordinator assembled the manuscript and checked repository integration; its exact variant and reasoning configuration were not exposed. No human mathematical contribution, independent human proof check or formal proof certification is supplied. Complete derivations and the consulted source references are included here; private reasoning transcripts are not published.

Final status and first remaining gap.

UNRESOLVED. Construct a continuum fixed-number density-preserving trial state whose additional kinetic and interaction costs are negligible in the source scaling.

Completion Estimate

COMPLETION: 0% This is an elementary auxiliary calculation or an explicit test of a restricted strategy. No new portion of the full source target is established.

Sources

• S1 ULAM.AI and the UnsolvedMath Contributors, UnsolvedMath, record 30005744 (OWR-14298014-003). Catalogue curation under CC BY 4.0; definitions and English exposition added locally. https://huggingface.co/datasets/ulamai/UnsolvedMath.
• S2 Mini-Workshop: Mathematics of Many-body Fermionic Systems $2023$, Oberwolfach Reports, DOI 10.4171/owr/2023/49, pp. 2809--2849. Original report and mathematical context. https://doi.org/10.4171/owr/2023/49.