Phase 6: Implement application validity and usefulness reports #505

Open
opened 2026-10-20 15:05:12 +02:00 by erikinkinen · 0 comments
Owner

Plan ID: P6-25
Record type: Implementation leaf
Scientific protocol: 2.2 · Backlog revision: 2.2.1
Specification: sections 10.10; common contracts in sections 3–5, 11 and 17.4.
Source: AES_PHASES_2_6_PLAN.md, issue heading at line 8895.
Plan SHA-256: 8acdaeb52d6a94fcad7de64d64521cd465f608344d63560b45f9202f5af71212
Protocol snapshot: Download the complete governing plan.

Goal

State what the held-out evidence can and cannot establish.

Prerequisites

Steps

  • Generate per-case workload support, task success, denial, alarm/hour and wall-clock lead reports.
  • Compare frozen observable support with training without retraining or removing adverse cases.
  • Implement conditional-AP versus pooled-AP labeling and all scope restrictions on forecasting claims.

Definition of Done

A persistent-failure forecast is not relabeled prevention or pre-first-violation prediction.

Execution and acceptance contract

Apply the governing semantic/data/statistical contracts in sections 3–5 and the relevant phase section, plus the section-11 checks applicable to this task. Section 17.4 defines the required acceptance evidence. Store the outcome and immutable prerequisite/input/output hashes at

artifacts/backlog/2.2.1/P6-25/acceptance.json.

Only explicit dependency and parent-closure edges govern readiness; numbering adds no predecessor dependency. Checkboxes are work/acceptance items, not commit quotas. Keep implementation changes independently reviewable using the existing PR template. Execution, review and gate tasks close with their evidence/decision records and a PR only when tracked artifacts change.

An unsupported scientific effect is a valid result. Missing correctness validation, invalid inference or absent required review is not completion. Do not change endpoints, matrices, margins, inference rules or sample sizes outside the protocol change-control process.

<!-- aes-plan-id: P6-25; backlog-revision: 2.2.1 --> **Plan ID:** `P6-25` **Record type:** Implementation leaf **Scientific protocol:** 2.2 · **Backlog revision:** 2.2.1 **Specification:** sections 10.10; common contracts in sections 3–5, 11 and 17.4. **Source:** `AES_PHASES_2_6_PLAN.md`, issue heading at line 8895. **Plan SHA-256:** `8acdaeb52d6a94fcad7de64d64521cd465f608344d63560b45f9202f5af71212` **Protocol snapshot:** [Download the complete governing plan](https://git.erikinkinen.fi/attachments/31cb768b-4788-4e38-aba9-9d7f0d1ee16d). ## Goal State what the held-out evidence can and cannot establish. ## Prerequisites - [P5-32 (#427)](https://git.erikinkinen.fi/erikinkinen/AES/issues/427) - [P6-21 (#491)](https://git.erikinkinen.fi/erikinkinen/AES/issues/491) ## Steps - [ ] Generate per-case workload support, task success, denial, alarm/hour and wall-clock lead reports. - [ ] Compare frozen observable support with training without retraining or removing adverse cases. - [ ] Implement conditional-AP versus pooled-AP labeling and all scope restrictions on forecasting claims. ## Definition of Done A persistent-failure forecast is not relabeled prevention or pre-first-violation prediction. ## Execution and acceptance contract Apply the governing semantic/data/statistical contracts in sections 3–5 and the relevant phase section, plus the section-11 checks applicable to this task. Section 17.4 defines the required acceptance evidence. Store the outcome and immutable prerequisite/input/output hashes at `artifacts/backlog/2.2.1/P6-25/acceptance.json`. Only explicit dependency and parent-closure edges govern readiness; numbering adds no predecessor dependency. Checkboxes are work/acceptance items, not commit quotas. Keep implementation changes independently reviewable using the existing PR template. Execution, review and gate tasks close with their evidence/decision records and a PR only when tracked artifacts change. An unsupported scientific effect is a valid result. Missing correctness validation, invalid inference or absent required review is not completion. Do not change endpoints, matrices, margins, inference rules or sample sizes outside the protocol change-control process.
erikinkinen added this to the Phase 6 milestone 2026-10-20 15:05:12 +02:00
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Reference
erikinkinen/AES#505
No description provided.