From 6e6e0fa1146443e7bea9c91d09fcc11c1f09712f Mon Sep 17 00:00:00 2001 From: jgrusewski Date: Fri, 1 May 2026 19:54:02 +0200 Subject: [PATCH] docs(sp5): remove duplicate Layer A close-out section (superseded by Layer D) --- ...nitude-differentiation-and-eval-collapse-design.md | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/docs/superpowers/specs/2026-05-01-sp5-magnitude-differentiation-and-eval-collapse-design.md b/docs/superpowers/specs/2026-05-01-sp5-magnitude-differentiation-and-eval-collapse-design.md index f287ccd1d..a508549c7 100644 --- a/docs/superpowers/specs/2026-05-01-sp5-magnitude-differentiation-and-eval-collapse-design.md +++ b/docs/superpowers/specs/2026-05-01-sp5-magnitude-differentiation-and-eval-collapse-design.md @@ -138,17 +138,6 @@ Code audit of `experience_action_select`'s Thompson direction-branch sampling: u Documented here for completeness; the SP5 brainstorm flagged this as a question. Code inspection resolved it. -### Layer A close-out: host-EMA → GPU sweep migration - -Migrate 5 remaining host-aggregated EMA sites to GPU per `feedback_no_cpu_compute_strict`: -- `training_sharpe_ema` — requires per-step PnL aggregation on GPU -- `max_dd_ema` — same -- `gamma_blend` — host-aggregated eval result, port to GPU -- `LearningHealth` composition pipeline — multi-step host arithmetic, port to GPU -- `HealthEmaTrackers` — q_gap_ema + q_var_ema + grad_norm_ema host-side EMAs - -Substantial: requires per-step trade-PnL aggregation kernel + multi-step composition kernel. Estimated 600-1000 LOC for this layer alone. - ## Architecture (mirror SP4 layer structure) ### Layer A — additive infrastructure