Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
75 changes: 75 additions & 0 deletions tracks/mps/solutions/Wander-276/PR_BODY.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,75 @@
<!-- wander-artwork -->
![Wander — Issue #276](https://raw.githubusercontent.com/JunkaiWang-TheoPhy/quantum.harness/664506886413b68cdbc5a362a6075c78fd5d2c46/docs/showcase/ranger-archive/assets/missions-v2/07-quantum-geometry-curved-state-space-v2.png)

> **EN** The spectrum is silent; the geometry is not.
>
> **中文** 能谱沉默,几何仍然说话。
<!-- /wander-artwork -->

## Team

| Field | Value |
|---|---|
| **Team name** | Wander |
| **Members** | Chenxi Wan, Yedi Shen, Junkai Wang |
| **Contact email** | WangTheoPhys@outlook.com |

## Challenge

Addresses #276: **What probes quantum chaos inside an exactly degenerate eigenspace?**

This PR gives an executable answer: the probe is the **non-Abelian quantum geometry of the protected projector over coupling space**. Exact degeneracy silences internal level statistics, while the response amplitudes (X_a=(1-P)\partial_aP) generate the quantum metric, Berry curvature, gauge-invariant four-channel tensors, and global holonomy.

## Scientific Advance

| Layer | Result |
|---|---|
| Exact geometry | (\mathcal Q_{ab}=X_a^\dagger X_b), with metric and Berry curvature as its symmetric and antisymmetric parts |
| Independent mechanism | Generic cubic (\mathcal N=2) SYK supplies a charge-resolved cohomological BPS manifold independent of the Kapit–Mueller/Laughlin parent |
| Exact structure | Nilpotency produces orthogonal exact/coexact response branches (X=X_-\oplus X_+), (X_-^\dagger X_+=0) |
| Sealed test | The complete $N=8,10,12$ pilot and separately sealed, held-out $N=14$ four-channel test reject both registered separable covariance nulls |
| Controls | Decomposable-supercharge curvature atoms and a one-sided Laughlin Gaussian-null regression |

The new scientific question is sharper than “is the Hamiltonian chaotic?” It asks whether the protected response tensor is controlled by registered two-point covariance data, or whether gauge-invariant four-point memory survives after that matching.

## Outcome-Blind Protocol

The held-out calculation freezes two parameter-free prediction models before opening the outcome sidecars: a collapsed separable covariance null and an exact/coexact Hodge-resolved separable covariance null. Complete disorder realizations—not tangent entries or tensor components—are the uncertainty unit. No scheduler dependency invokes unsealing; seal validation and scoring require a separate explicit action.

**Final held-out verdict:** independent seal validation passed before explicit unsealing, and the frozen branch is `cohomological_non_gaussian_class`.

| $N=14$ sparse sector | Physical median (95% bootstrap) | Collapsed null (97.5% prediction) | Hodge null (97.5% prediction) |
|---|---:|---:|---:|
| Adjacent | 0.301529 [0.291527, 0.312061] | [0.111789, 0.111852] | [0.112344, 0.112513] |
| Central | 0.374993 [0.368980, 0.380473] | [0.111338, 0.111353] | [0.111333, 0.111348] |

Both primary physical intervals are disjoint from both sealed prediction intervals. The prediction JSON was frozen at SHA-256 `fc300dc7e4bdc1be157919e458ac868d3468533cce31108f23c9fba4f7e9f102`; the independently recomputable inference artifact is `177643e07fc6cf210362fc1077070bd1f0ba316b6805042a626de3f96c55a627`.

## Claim Boundary

- Established: an independent supersymmetric protection mechanism, exact Hodge-resolved response identity, analytic controls, and finite-size $N=8$--$14$ response memory beyond both frozen separable covariance-null families.
- Not established: asymptotic Geometric ETH, thermalization, real-time chaos, or a thermodynamic-limit theorem.
- The selected null-rejection branch means memory beyond the frozen separable Hodge/collapsed covariance models; it does not prove non-Gaussianity relative to every possible entrywise covariance model.

## Deliverables

- [Harness solution overview](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/README.md)
- [Research and reproduction guide](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/README.md)
- [Letter source](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/overleaf_sync/cohomological_geometric_eth/main.tex)
- [Supplement source](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/overleaf_sync/cohomological_geometric_eth/supplement.tex)
- [Compiled Letter](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/01_task_folder/task_05/script/output/response_complex_memory_v7.pdf)
- [Compiled Supplemental Material](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/01_task_folder/task_05/script/output/response_complex_memory_supplement_v7.pdf)
- [Machine-generated result report](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/01_task_folder/task_05/script/output/susy_hodge_geometric_eth_report_v7.md)
- [Scientific ceiling and literature memo](https://github.com/JunkaiWang-TheoPhy/quantum.harness/blob/codex/issue-276-quantum-geometry/tracks/mps/solutions/Wander-276/research/docs/2026-08-01-scientific-ceiling-strategy.md)
- [Public research branch](https://github.com/JunkaiWang-TheoPhy/Chaos-of-Quantum-Geometry/tree/codex/task-05-geometric-chaos-baseline)

## Verification

```bash
cd tracks/mps/solutions/Wander-276
bash verify.sh
```

The final verifier checks exact identities, controls, the complete pilot and held-out grids, the seal state machine, every compact hash, frozen branch recomputation, figure provenance, byte-exact Letter/Supplement PDFs, and 43 focused v7 tests.

@OkongOyangO, please review the protected-response interpretation, the exact/coexact identity, and the outcome-blind four-channel test as Wander's Issue #276 submission.
90 changes: 90 additions & 0 deletions tracks/mps/solutions/Wander-276/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,90 @@
# Wander — Issue #276: Chaos in Exactly Degenerate Quantum Manifolds

![Quantum geometry over an exactly degenerate manifold](https://raw.githubusercontent.com/JunkaiWang-TheoPhy/quantum.harness/664506886413b68cdbc5a362a6075c78fd5d2c46/docs/showcase/ranger-archive/assets/missions-v2/07-quantum-geometry-curved-state-space-v2.png)

> **EN** The spectrum is silent; the geometry is not.
>
> **中文** 能谱沉默,几何仍然说话。

## Team

| Field | Value |
|---|---|
| **Team name** | Wander |
| **Members** | Chenxi Wan, Yedi Shen, Junkai Wang |
| **Contact email** | WangTheoPhys@outlook.com |

## Challenge and Answer

This submission addresses [QuantumBFS/quantum.harness#276](https://github.com/QuantumBFS/quantum.harness/issues/276): **What probes quantum chaos inside an exactly degenerate eigenspace?**

The executable answer is the **non-Abelian quantum geometry of the protected projector over coupling space**. If (P(\lambda)) is the degenerate projector and (X_a=(1-P)\partial_aP), then

$$\mathcal Q_{ab}=X_a^\dagger X_b,\qquad g_{ab}=\frac{\mathcal Q_{ab}+\mathcal Q_{ba}}{2},\qquad F_{ab}=i(\mathcal Q_{ab}-\mathcal Q_{ba}).$$

Ordinary level statistics is undefined inside a flat multiplet, but (g), (F), gauge-invariant four-channel contractions, Chern numbers, and Wilson holonomy remain nontrivial.

## Independent Protection Mechanisms

The release no longer rests on one fractional-quantum-Hall example.

| Mechanism | Protected manifold | Exact response structure | Role |
|---|---|---|---|
| Kapit–Mueller/Laughlin parent (H=B^\dagger B) | Frustration-free, gapped zero modes | One-sided response (X=-H_\perp^+B^\dagger\delta BP) | Local topological benchmark, Jacobi curvature, fixed-Chern holonomy |
| Generic cubic (\mathcal N=2) SYK (H=\{Q,Q^\dagger\}) | Charge-resolved harmonic cohomology | Orthogonal exact/coexact response (X=X_-\oplus X_+) | Independent supersymmetric mechanism and sealed covariance test |

Nilpotency gives (X_-^\dagger X_+=0) exactly. The two-sided Hodge decomposition is therefore a physical response-complex structure, not a basis rewrite of the Laughlin calculation.

## Registered Scientific Test

For eight coupling tangents, the code whitens the channel covariance and evaluates the gauge-invariant tensor

$$\mathcal T_{abcd}=\frac1D\operatorname{Tr}(\widehat X_a^\dagger\widehat X_b\widehat X_c^\dagger\widehat X_d).$$

The (N=8,10,12) pilot uses complete disorder realizations as the uncertainty unit. A held-out (N=14) calculation freezes two prediction models before any four-channel outcome is opened:

1. a collapsed separable covariance Gaussian null;
2. an exact/coexact Hodge-resolved separable covariance Gaussian null.

Safe covariates and predictions are SHA-256 sealed. Unsealing is an explicit, separate command and is never launched by the scheduler dependency chain. The seal was independently checked before opening, and the frozen result is:

| $N=14$ sparse sector | Physical median (95% bootstrap) | Collapsed null (97.5% prediction) | Hodge null (97.5% prediction) |
|---|---:|---:|---:|
| Adjacent | 0.301529 [0.291527, 0.312061] | [0.111789, 0.111852] | [0.112344, 0.112513] |
| Central | 0.374993 [0.368980, 0.380473] | [0.111338, 0.111353] | [0.111333, 0.111348] |

Both nulls miss both primary sectors, selecting `cohomological_non_gaussian_class`.

## Claim Boundary

This submission establishes an **independent model/operator class** for protected quantum-geometric chaos and finite-size $N=8$--$14$ four-channel response memory beyond two preregistered separable covariance nulls. It does **not** claim asymptotic Geometric ETH, conventional energy-resolved ETH, real-time chaos, or a thermodynamic-limit theorem. Null rejection does not by itself prove intrinsic non-Gaussianity after matching every entrywise nonseparable covariance.

## Deliverables

- [Research and reproduction guide](research/README.md)
- [Task-level guide](research/01_task_folder/task_05/README.md)
- [Letter source](research/overleaf_sync/cohomological_geometric_eth/main.tex)
- [Supplement source](research/overleaf_sync/cohomological_geometric_eth/supplement.tex)
- [Compiled Letter](research/01_task_folder/task_05/script/output/response_complex_memory_v7.pdf)
- [Compiled Supplemental Material](research/01_task_folder/task_05/script/output/response_complex_memory_supplement_v7.pdf)
- [Exact-data result report](research/01_task_folder/task_05/script/output/susy_hodge_geometric_eth_report_v7.md)
- [Scientific ceiling and literature memo](research/docs/2026-08-01-scientific-ceiling-strategy.md)
- [Public research branch](https://github.com/JunkaiWang-TheoPhy/Chaos-of-Quantum-Geometry/tree/codex/task-05-geometric-chaos-baseline)

## Verification

```bash
cd tracks/mps/solutions/Wander-276
bash verify.sh
```

The verification path checks the exact Hodge identity, analytic controls, pilot-grid completeness, outcome-blind sealing, artifact hashes, frozen branch selection, figure provenance, and byte-exact Letter/Supplement delivery.

## Next Scientific Gate

The finite-size result is designed to expose the next theorem rather than hide it. The two high-ceiling targets are:

- an asymptotic concentration/scaling law for the covariance-controlled response tensor;
- a spatially local nilpotent-supercharge model with a stable zero-mode count, open gap, and moving cohomological projector.

These are journal-ceiling gates, not assumptions built into the current claim.
Loading