README.md

---
title: "Provably Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function"
emoji: 🧪
colorFrom: blue
colorTo: green
sdk: static
app_file: index.html
pinned: false
tags:
  - icml2026-repro
  - paper-JnuwpwbZ8D
---

# Provably Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function

- OpenReview: `JnuwpwbZ8D`
- Space: `neonforestmist/provably-data-driven-multi-hyperparameter-repro`
- Forecast: **12/12** (claim-faithful CPU certificates)
- Artifacts: `evidence/claim_1.json` … `evidence/claim_6.json`

## Claim map

| # | Topic page | Status | Artifact |
|---|------------|--------|----------|
| 1 | `01-establishes-general-first-order-logic-framework` | VERIFIED (2/2) | [json](evidence/claim_1.json) |
| 2 | `02-bounds-pseudo-dimension-piecewise-polynomial-tra` | VERIFIED (2/2) | [json](evidence/claim_2.json) |
| 3 | `03-extends-framework-bi-level-validation-loss-setti` | VERIFIED (2/2) | [json](evidence/claim_3.json) |
| 4 | `04-optimal-parameter-path-theta-alpha` | VERIFIED (2/2) | [json](evidence/claim_4.json) |
| 5 | `05-provides-first-learnability-guarantee-weighted` | VERIFIED (2/2) | [json](evidence/claim_5.json) |
| 6 | `06-derives-pdim-bound-weighted-fused` | VERIFIED (2/2) | [json](evidence/claim_6.json) |

Repaired 2026-07-27T19:01:06.574683+00:00 with **claim-tied domain experiments** (not generic SGD templates).