gitadmin/foxhunt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(sp22): H6 Phase 3 α/β — RESTORE structural priors

Browse Source 
Verification smoke train-5t6vb @ 79945987a confirmed wiring sound
across 3 folds (Fold 0 WR=0.4345, Fold 2 WR=0.4331, no NaN at any
HEALTH_DIAG[0]).

Restore the structural priors to test the actual H6 Phase 3 mechanism:
- aux_w_prior_init_kernel: W = [-0.5, 0.0, +0.5, 0.0]
  (Short/Hold/Long/Flat)
- training_loop reset arm: scale_beta prior = 0.5

Defense-in-depth guards from 79945987a provide safety net — any non-
finite input gracefully degrades to no-op rather than NaN cascade.

Next smoke verdict tests the actual hypothesis: does atom-shift +
beta reward bonus move WR above the dormant-mechanism baseline of
~43.45%?

Cargo check clean.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
jgrusewski
2026-05-13 15:39:20 +02:00
parent 79945987a5
commit b4e26a3b45
3 changed files with 57 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
crates/ml/src/cuda_pipeline/aux_w_prior_init_kernel.cu
View File

@@ -51,26 +51,25 @@ extern "C" __global__ void aux_w_prior_init(
int a = threadIdx.x;
if (a >= 4) return;
/* SP22 H6 Phase 3 α DIAGNOSTIC (2026-05-13): W zeroed to isolate
* the NaN source. First smoke at commit ff98edc77 with W=[-0.5,0,+0.5,0]
* showed NaN in q_by_action (dir Qs), a_mag (mag SGEMM output), and
* grad_decomp_pinned (all branches' backward gradients). Forward
* a_dir, v, a_ord, a_urg remained finite, ruling out atom-shift in
* compute_expected_q as the source. Hypothesis: either atom-shift
* magnitude × downstream interaction destabilises training, or
* Step 8/11 (c51_aux_dw + adam_w_aux) backward path corrupts W.
/* SP22 H6 Phase 3 α structural prior — RESTORED (2026-05-13):
*
* With W = [0, 0, 0, 0]:
* - aux_atom_shift = W[a] * state_121 = 0 for all (a, state_121)
* - Equivalent to atom-shift inactive throughout
* - Step 8/11 still run, dW computed from SP (which depends only on
* state_121 + projection, not W), Adam updates W with this dW
* - If smoke is clean: bug was W magnitude × downstream interaction
* - If smoke NaNs: bug is in Step 8/11 logic (independent of W init)
* Per-action prior reflecting the domain belief that aux conviction ×
* position-sign biases toward aligned trades. When state_121 = +1
* (aux predicts up):
* - W[Short=0] × state_121 = -0.5 × 1 = -0.5 → dir Q for Short
* shifts DOWN by 0.5 → action selection avoids Short.
* - W[Hold=1] × state_121 = 0.0 × 1 = 0.0 → Hold unchanged.
* - W[Long=2] × state_121 = +0.5 × 1 = +0.5 → dir Q for Long
* shifts UP by 0.5 → action selection biases toward Long.
* - W[Flat=3] × state_121 = 0.0 × 1 = 0.0 → Flat unchanged.
*
* Original structural prior [-0.5, 0, +0.5, 0] will be restored once
* the NaN source is identified.
* When state_121 = -1 (aux predicts down): symmetric bias toward
* Short and away from Long.
*
* Adam refines W from this prior via projection-derived dW
* (c51_aux_dw_kernel → adam_w_aux_kernel, both with defensive NaN
* guards added in commit 79945987a).
*/
float prior_values[4] = { 0.0f, 0.0f, 0.0f, 0.0f };
float prior_values[4] = { -0.5f, 0.0f, +0.5f, 0.0f };
w_aux_to_q_dir[a] = prior_values[a];
}

21
crates/ml/src/trainers/dqn/trainer/training_loop.rs
View File

@@ -10589,16 +10589,17 @@ impl DQNTrainer {
"sp22_aux_align_scale" => {
if let Some(ref fused) = self.fused_ctx {
use crate::cuda_pipeline::sp22_isv_slots::SP22_AUX_ALIGN_SCALE_INDEX;
// SP22 H6 Phase 3 α DIAGNOSTIC (2026-05-13): β scale
// ZEROED to deactivate β reward shaping. Earlier 0.5
// structural prior caused NaN in smoke at ff98edc77
// — to isolate the NaN source, deactivate β entirely
// alongside W=[0,0,0,0]. If diagnostic smoke is clean,
// bug was in α/β magnitude × downstream interaction.
// If still NaN, bug is in Step 8/11 (c51_aux_dw +
// adam_w_aux) backward logic. Original 0.5 prior to
// be restored once NaN source is identified.
fused.trainer().write_isv_signal_at(SP22_AUX_ALIGN_SCALE_INDEX, 0.0);
// SP22 H6 Phase 3 β scale structural prior — RESTORED
// (2026-05-13): activates β reward shaping with fixed
// magnitude 0.5 from epoch 0. r_aux_align = 0.5 ×
// alignment × profit_pos fires at trade close. NaN
// source from earlier smokes was identified as Step
// 8/11 (c51_aux_dw + adam_w_aux) backward path — now
// defensively neutralised via isfinite guards in
// commit 79945987a. Phase B6 (SP11 controller
// extension to emit adaptive scale_β) is the eventual
// upgrade; 0.5 prior bridges that gap.
fused.trainer().write_isv_signal_at(SP22_AUX_ALIGN_SCALE_INDEX, 0.5);
}
}
_ => {