實驗EXP-2026-0011v0.1

PACC-Lab v0.4——相關來源與依賴幾何

來源分成叢集並帶潛在共同反轉；正確的參考用聯合概似，天真的獨立乘積 Bayes 是負控制（在中、高相關幾何下聯合勝過天真；獨立幾何下差距恰為 0）。非概率系統只看得到叢集歸屬。任務狀態收斂存活：N0/N1/N3 在中、高相關下通過（N2 在高相關下一致度掉到 0.8364）。潛在依賴座標——叢集處於受污染分支的後驗——任何系統的 train-only affine-sigmoid 映射都無法重建：真實 JS ≈ shuffled ≈ constant。

研究狀態: STABLE 目前的研究結論相對穩定
證據等級: E3 重複實驗
結果: MIXED
資料基礎: SYNTHETIC 合成數據與理論推理。現在很多人把合成數據當成真的；這個實驗室刻意反過來說——在真正的混合模型出現之前，推論就只是推論，理論上可能不等於實際上可能。
版本: 0.1
更新: 2026-09-09
建立: 2026-09-09
領域: Model Representation, Evaluation, Formal AI
計畫: PRG-2026-0001 自適應世界狀態系統的第一原理框架
作者: Neo.K (EveMissLab)
AI 協作: Sol (GPT-5.6, OpenAI ChatGPT)

假設

hypothesis: Task-state convergence survives dependent evidence, and the non-probabilistic relation state itself maps to the Bayesian common-cause posterior.

設定

model_ids: MOD-2026-0001 N0——帶號支持
MOD-2026-0002 N1——序數錦標賽
MOD-2026-0003 N2——帶號圖
MOD-2026-0004 N3——約束競爭
MOD-2026-0005 Bayesian 參考（精確後驗／Beta-Bernoulli／聯合共同因／HMM）
dataset_ids: DAT-2026-0001 PACC 合成證據世界
benchmark_ids: BEN-2026-0001 PACC 微型實驗室協定 v0.1（凍結門檻）
software_environment: Python; deterministic seeded generators; no network, no LLM.

執行

run_count

random_seeds

20260909
6 secondary seeds
4 high-correlation seeds at primary scale

controls

shuffled-target mapping
constant prediction
broken composition (v0.6+)
target-local refit (v0.7+)

metrics

verdict

TASK-LEVEL CONVERGENCE SURVIVES DEPENDENT EVIDENCE; LATENT DEPENDENCE-STATE CONVERGENCE NOT SUPPORTED

environment

joint_vs_naive_logloss

independent: 1.245687
1.245687
moderate: 1.632599
1.834445
high: 1.758082
2.018617

task

moderate

N0

agreement: 0.955247
D_R: 0.008095

N3

agreement: 0.942901
D_R: 0.008234

high

N0

agreement: 0.861111
D_R: 0.009827

N2

agreement: 0.83642
D_R: 0.017874

dependence_coordinate

moderate_real_js: 0.0605
moderate_shuffled: 0.0641
moderate_constant: 0.0615
high_real_js: 0.0398
high_shuffled: 0.04
high_constant: 0.0397
pass: 0

詮釋

interpretation: A layered picture: a decision-relevant quotient state can converge to a probabilistic coordinate while a deeper latent explanatory variable stays representation-dependent — evidence against the strongest 'everything becomes probability-like' reading.

限制

limitations: A result about the current state representation and frozen readout, not a theorem that no richer relation graph could encode the latent variable.
Cluster membership is given, not learned.

重現

reproduction_instructions: Extract the version's FINAL bundle; python -m pytest -q; run the version's primary script with the recorded seed; docs/PACC_LAB_v0.N_RESULTS.md and docs/EXPERIMENT_PROTOCOL_v0.N.md are inside the bundle.

結果

來源	關係	目標	狀態	ID
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`produces`	`RST-2026-0008` v0.4：潛在依賴座標無法重建	`ACTIVE`	`REL-2026-0204`

記錄欄位

completed_at: 2026-09-09

關係

來源	關係	目標	狀態	ID
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`runs_on`	`SYS-2026-0002` PACC-Lab——收斂猜想的微型實驗室	`ACTIVE`	`REL-2026-0193`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_benchmark`	`BEN-2026-0001` PACC 微型實驗室協定 v0.1（凍結門檻）	`ACTIVE`	`REL-2026-0194`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_dataset`	`DAT-2026-0001` PACC 合成證據世界	`ACTIVE`	`REL-2026-0195`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_model`	`MOD-2026-0001` N0——帶號支持	`ACTIVE`	`REL-2026-0196`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_model`	`MOD-2026-0002` N1——序數錦標賽	`ACTIVE`	`REL-2026-0197`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_model`	`MOD-2026-0003` N2——帶號圖	`ACTIVE`	`REL-2026-0198`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_model`	`MOD-2026-0004` N3——約束競爭	`ACTIVE`	`REL-2026-0199`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`uses_model`	`MOD-2026-0005` Bayesian 參考（精確後驗／Beta-Bernoulli／聯合共同因／HMM）	`ACTIVE`	`REL-2026-0200`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`tests`	`THY-2026-0005` PACC 猜想——四層收斂階梯	`ACTIVE`	`REL-2026-0201`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`extends`	`EXP-2026-0010` PACC-Lab v0.3——無 oracle 的來源可靠度學習	`ACTIVE`	`REL-2026-0202`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`produced`	`ART-2026-0024` PACC-Lab v0.4 Correlated Sources FINAL `artifact://evemisslab/adaptive-epistemic-systems/PACC-Lab_v0.4_Correlated_Sources_FINAL.zip`	`ACTIVE`	`REL-2026-0203`
`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`produces`	`RST-2026-0008` v0.4：潛在依賴座標無法重建	`ACTIVE`	`REL-2026-0204`
`EXP-2026-0012` PACC-Lab v0.5——更豐富的關係狀態能救回潛在座標嗎？	`extends`	`EXP-2026-0011` PACC-Lab v0.4——相關來源與依賴幾何	`ACTIVE`	`REL-2026-0215`

歷史與來源歷程

Canonical URL

https://evemisslab.com/ai/experiments/EXP-2026-0011/

機器可讀

/ai/experiments/EXP-2026-0011/index.json

快照

AI-SNAPSHOT-v0.1-fe85b9694a45

來源歷程

source: EveMissLab research collection: Adaptive Epistemic Systems (真本體論13)
extracted_by: Splice (Claude Code), reading the canonical UTF-8 sources and each lab's own result reports
extracted_at: 2026-09-11
generator: tools/extract_aes/extract.py
claim_boundary: status, evidence level and result type follow the source artifact's own stated claim boundary; nothing is upgraded beyond what the report supports