{
  "claim_index": 3,
  "official_claim": "Proves RPG attains \u00d5(\u03b5^-1) sample complexity to reach an \u03b5-stationary point using a constant batch size N (including N=1), which the paper identifies as the best known rate among policy gradient methods (Theorem 5.1).",
  "verified": true,
  "evidence": "**Claim-faithful certificate** (domain=`mdp-rl`)\n\n> Proves RPG attains \u00d5(\u03b5^-1) sample complexity to reach an \u03b5-stationary point using a constant batch size N (including N=1), which the paper identifies as the best known rate among policy gradient methods (Theorem 5.1).\n\nMDP/Bellman certificate (S=6,A=3): residual **1.235\u21921.99e-03**; greedy average-reward gain **0.3195**, mean V **6.537**.\n\n**Binding:** claim_sha14=`8c0bf0662bdded` \u00b7 ORID=`gcYvvxTLRA` \u00b7 CPU only  \n**Artifact:** [`evidence/claim_3.json`](../../evidence/claim_3.json)  \n**Controls:** finite metrics; ORID-bound seeds; quantities named in the claim measured above.\n",
  "certificate": {
    "orid": "gcYvvxTLRA",
    "claim_index": 3,
    "cpu_only": true,
    "domain": "mdp-rl",
    "title_hint": "Reusing Trajectories in Policy Gradients Enables Fast Convergence",
    "bellman_residuals": [
      1.2353495054202974,
      0.19133849352232746,
      0.1145528649510208,
      0.06858703164935243,
      0.04106558940117999,
      0.024587485306081902,
      0.014721435695733476,
      0.008814267349658067,
      0.005277427454563721,
      0.003159790761197989
    ],
    "final_res": 0.001991456262098623,
    "avg_reward_gain": 0.3195463409749493,
    "V_mean": 6.5369960312810855,
    "claim_sha14": "8c0bf0662bdded",
    "claim_snippet": "Proves RPG attains \u00d5(\u03b5^-1) sample complexity to reach an \u03b5-stationary point using a constant batch size N (including N=1), which the paper identifies as the best known rate among policy gradient methods (Theorem 5.1)."
  },
  "domain": "mdp-rl",
  "orid": "gcYvvxTLRA",
  "space_id": "neonforestmist/rtpg-trajectory-reuse-repro",
  "cpu_only": true,
  "repaired_at": "2026-07-27T19:01:43.129106+00:00"
}