{
  "schema_version": 1,
  "title": "Optimal Regularization for Performative Learning",
  "emoji": "🎯",
  "space_id": "SabaPivot/repro-optimal-regularization-for-performative-learning",
  "paper": {
    "arxiv_id": "2510.12249"
  },
  "tags": [
    "icml2026-repro",
    "paper-G4ve69pimc"
  ],
  "updated_at": "2026-07-31T08:54:09.630153+00:00",
  "root": {
    "slug": "index",
    "title": "Optimal Regularization for Performative Learning",
    "file": "pages/index.md",
    "children": [
      {
        "slug": "executive-summary",
        "title": "Executive summary",
        "file": "pages/executive-summary/page.md",
        "children": []
      },
      {
        "slug": "claim-1-population-excess-risk-characterization",
        "title": "In the population setting, Theorem 1 characterizes excess risk as a function of the magnitude and direction of the performative effect together with spurious features (Section 4, Theorem 1).",
        "file": "pages/claim-1-population-excess-risk-characterization/page.md",
        "children": []
      },
      {
        "slug": "claim-2-optimal-regularization-proportional",
        "title": "Corollary 2 shows the optimal regularization parameter in the population regime is proportional to the strength of the performative effect, with optimal risk remaining strictly positive (Section 4, Corollary 2).",
        "file": "pages/claim-2-optimal-regularization-proportional/page.md",
        "children": []
      },
      {
        "slug": "claim-3-deterministic-equivalent-fixed-point",
        "title": "Theorem 3 establishes a deterministic equivalent of the performative fixed point for over-parameterized ridge regression when the number of features exceeds the number of samples (Section 5, Theorem 3).",
        "file": "pages/claim-3-deterministic-equivalent-fixed-point/page.md",
        "children": []
      },
      {
        "slug": "claim-4-optimal-regularization-sign-flip-noise",
        "title": "Theorem 4 shows the optimal regularization moves in the same direction as the performative effect on predictive features under low noise, but in the opposite direction under high noise, in the over-parameterized regime (Section 5, Theorem 4).",
        "file": "pages/claim-4-optimal-regularization-sign-flip-noise/page.md",
        "children": []
      },
      {
        "slug": "claim-5-overparam-performativity-improves-risk",
        "title": "Numerical experiments in Section 6 confirm that in the over-parameterized setting, performative effects can improve optimally-regularized risk when performativity reinforces existing trends, contrasting with the population-regime degradation (Section 6).",
        "file": "pages/claim-5-overparam-performativity-improves-risk/page.md",
        "children": []
      },
      {
        "slug": "claim-6-methods-provenance",
        "title": "Methods, independence & provenance",
        "file": "pages/claim-6-methods-provenance/page.md",
        "children": []
      },
      {
        "slug": "claim-7-failure-boundaries",
        "title": "Failure boundaries & honest scope",
        "file": "pages/claim-7-failure-boundaries/page.md",
        "children": []
      },
      {
        "slug": "claim-99-fresh-independent-cpu-audit",
        "title": "Fresh independent CPU audit",
        "file": "pages/claim-99-fresh-independent-cpu-audit/page.md",
        "children": []
      },
      {
        "slug": "conclusion",
        "title": "Conclusion",
        "file": "pages/conclusion/page.md",
        "children": []
      }
    ]
  },
  "agent_view_tokens": 11088,
  "revision": "1785253564933160000"
}