jgrusewski 4249ebc961 feat(sp20): register 10 ISV slots in StateResetRegistry ...

Sentinel = 0.0 per pearl_first_observation_bootstrap. First observation
of each EMA replaces the sentinel directly, no blending.

Slots [510..520):
- loss_cap (510): adaptive loss cap for reward clamp
- alpha_ema (511): Wiener-α EMA for loss_cap producer
- wr_ema (512): win-rate EMA driving loss_cap adaptive ramp
- hold_cost_scale (513): hold penalty cost multiplier
- target_hold_pct (514): hold-engagement target
- hold_pct_ema (515): hold-engagement EMA
- hold_reward_ema (516): hold-action reward EMA
- n_step (517): multi-step TD horizon adapter
- aux_conf_threshold (518): auxiliary task confidence threshold
- aux_gate_temp (519): auxiliary task gating temperature

2026-05-09 18:17:38 +02:00

..

infra

feat(infra): GPU training launcher with local/cloud routing

2026-02-23 11:19:58 +01:00

plans

feat(dqn): enable Branching DQN with 45 factored actions (5×3×3)

2026-03-11 22:00:13 +01:00

superpowers

plan(sp19+20): WR-first reward implementation plan

2026-05-09 18:04:00 +02:00

dqn-backward-nan-audit.md

fix(dqn): SP3 Mech 10 — ISV-driven h_s2 activation clamp + slot 49 diag

2026-04-30 17:26:25 +02:00

dqn-gpu-hot-path-audit.md

fix(sp15-cubin-preload): pre-load 6 SP15 launchers' CudaFunction handles in experience collector

2026-05-07 23:36:07 +02:00

dqn-named-dims.md

fix(dqn): mag_concat_qdir SH2+b0 vs SH2+3 off-by-one — buffer + weight migration

2026-04-27 22:19:41 +02:00

dqn-wire-up-audit.md

feat(sp19 commit b): producer-side multi-horizon reward blend at fxcache write time

2026-05-09 14:45:20 +02:00

h_s2_consumers_audit.md

audit(dqn-v2): h_s2 audit anchor corrections (post-2c.3+4 reconnaissance)

2026-04-25 12:29:47 +02:00

health_diag_inventory.md

feat(diag): Phase 2A — health_diag_isv_mirror kernel + GpuHealthDiag orchestrator

2026-04-28 23:54:45 +02:00

isv-slots.md

feat(sp20): register 10 ISV slots in StateResetRegistry

2026-05-09 18:17:38 +02:00

lookahead-bias-audit-2026-04-28.md

fix(audit): lookahead housekeeping — purge gap + per-fold NormStats

2026-05-09 13:52:00 +02:00

ml-supervised-to-dqn-concept-audit.md

plan4(task7): Part E audit close-out — every supervised concept landed or OUT

2026-04-26 10:36:03 +02:00

sp18-wireup-audit.md

chore(sp18): refresh audit fingerprint after Phase 4 docstring update

2026-05-09 11:26:17 +02:00