diff --git a/crates/ml-alpha/src/trainer/integrated.rs b/crates/ml-alpha/src/trainer/integrated.rs index 84d7bd35f..8a5bda85f 100644 --- a/crates/ml-alpha/src/trainer/integrated.rs +++ b/crates/ml-alpha/src/trainer/integrated.rs @@ -3059,6 +3059,25 @@ impl IntegratedTrainer { (crate::rl::isv_slots::RL_REWARD_CLAMP_CLIP_RATE_EMA_INDEX, 0.0_f32), // v9 — arm the defensive warmup window. (crate::rl::isv_slots::RL_EVAL_WARMUP_REMAINING_INDEX, warmup_steps), + // Regime observer boundary policy (F1.7 — spec 2026-05-31 v3, + // [[pearl_adaptive_carryover_discipline]]): + // - Transient state (FLAG/DURATION/TIMEOUT) → 0 + // - RECOVERY_FACTOR → 1.0 (no recent tail; ε = ε_max) + // - TAIL_EVENT_RECENCY → 1e6 sentinel ("no tails seen yet") + // - KELLY_EPS_RECOVERY_LIVE → 0.50 (= ε_max default) + // - PREV_WORST_PNL → 0 (CRITICAL — must align with the just-reset + // session_pnl_worst above to prevent spurious 3σ tail event) + // - POPART_MAX_ABS_REWARD_EMA → 0 (F4 envelope reset) + // - Welford state (M2/MEAN/COUNT) intentionally NOT reset — the + // variance of session_pnl_change is regime-invariant. + (crate::rl::isv_slots::RL_REGIME_DEAD_ZONE_FLAG_INDEX, 0.0_f32), + (crate::rl::isv_slots::RL_REGIME_DEAD_ZONE_DURATION_INDEX, 0.0_f32), + (crate::rl::isv_slots::RL_REGIME_DEAD_ZONE_TIMEOUT_FLAG_INDEX, 0.0_f32), + (crate::rl::isv_slots::RL_REGIME_TAIL_EVENT_RECENCY_INDEX, 1.0e6_f32), + (crate::rl::isv_slots::RL_REGIME_RECOVERY_FACTOR_INDEX, 1.0_f32), + (crate::rl::isv_slots::RL_KELLY_EPS_RECOVERY_LIVE_INDEX, 0.50_f32), + (crate::rl::isv_slots::RL_REGIME_PREV_WORST_PNL_INDEX, 0.0_f32), + (crate::rl::isv_slots::RL_POPART_MAX_ABS_REWARD_EMA_INDEX, 0.0_f32), ] { let mut args = RawArgs::new(); args.push_ptr(self.isv_dev_ptr);