Observed train-vpb4w freeze at epoch 23 (val_Sharpe=51.11):
- avg_grad=10.0 (nonzero!) but Q-stats FROZEN for 4+ epochs
- Root cause: Adam momentum converged to fixed point, not zero gradient
Component 10: Targeted Adam momentum reset for branch weights only
(indices 8-41). Lighter than full rewind, preserves trunk convergence.
Component 11: Q-gap target attractor for spread gradient. Amplifies
spread when Q-gap is below target, reduces when above. Prevents
spread from being too weak at high Q-gap levels.
Component 12: Cosine LR schedule with warm restarts (T=20 epochs).
LR decay prevents overshoot, restart jumps break fixed points.
Half-momentum at restart preserves gradient direction.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Task 10: q_mean_reduce + q_mean_subtract two-phase kernel. Zero params.
Fixes Q-mean drift 0→+0.47 from bootstrapping bias.
Task 11: Adaptive entropy_coeff based on utilization_ema. Zero params.
Fixes atom utilization collapse 100%→20%.
Execution order: 10, 11, then 1-9.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Observed in train-vpb4w (live H100 run):
- Q-mean drifts 0→+0.47 over 18 epochs (bootstrapping bias)
- Atom utilization drops 100%→20% (distributional collapse)
- val_Sharpe peaks 54.59 then decays to 31.52
Component 8: Mean-center Q-values before action selection pipeline.
Zero params, preserves ranking, prevents drift accumulation.
Component 9: Adaptive atom entropy regularization. Ramps entropy_coeff
when utilization drops below 50%. Uses existing c51_grad_kernel param.
Both are zero-param fixes moved to top of implementation order.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Checkpoint ring buffer ranked by improvement rate, plateau detection
via liquid tau velocity, informed perturbation cycle (Adam reset,
shrink-perturb, temp boost, LR×2), depth-limited rewind. ~150 lines
in training_loop.rs, zero CUDA changes. Recovers from plateaus that
prevention mechanisms (spread gradient, liquid tau) fail to avoid.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Raw E[Q] ~ 0.1 while h_s2 ~ 1.0 post-ReLU. Without normalization,
Q_dir columns learn 10x slower. Dividing by delta_z scales to ~1-10.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Magnitude FC input changes from h_s2[B, SH2] to [h_s2; Q_dir][B, SH2+3].
Direction E[Q] values (3 scalars per sample) are computed from branch 0
logits and concatenated. w_b1fc grows by 3*adv_h parameters (768 for AH=256).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
6 new CUDA kernels, C51/grad/epsilon kernel mods, dual-tau IQL,
v_range dead code deletion, per-sample everything.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Covers: PER advantage-weighted staleness, per-sample C51 atom
support centered on V(s), per-branch advantage decomposition,
dual-tau expectile gap exploration, and v_range dead code removal.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fundamental fix: C51 atom spacing (dz=0.6) was larger than the Q-value
range (0.02), making the distributional loss unable to resolve rewards.
The network was blind to its own learning signal.
Adaptive v_range via device buffer (same pattern as tau_buf):
- v_range_buf[2] on GPU, all C51/MSE/CQL/expected_q kernels read via
pointer (graph captures stable address, reads value at replay time)
- Cold start: v_range=±1.0 → dz=2/51=0.039 → 25× atom resolution
- Monotonic expansion based on Q-stats every 50 training steps
- Never shrinks (avoids invalidating Bellman targets in replay buffer)
Kernel changes (5 files):
- c51_loss_kernel.cu, mse_loss_kernel.cu, mse_grad_kernel.cu,
cql_grad_kernel.cu, compute_expected_q: scalar v_min/v_max → pointer
Also fixed:
- PopArt warmup 100→1 (start normalizing from step 1)
- PopArt variance floor 0.0001→1e-8 (allow sigma < 0.01)
- Backtest evaluator: allocate own v_range_buf for compute_expected_q
- num_atoms 51→52 in all TOMLs (TF32 4-element alignment)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
DuelingWeightSet becomes raw u64 pointers into params_buf, not separate
CudaSlice allocations. Eliminates ALL D2D copies between weight
representations. flatten/unflatten cycle deleted entirely.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
VarStore is a CPU-trip corrupting data + degrading performance. Must be
replaced with direct pointer buffers (flat params_buf is already the
source of truth). Checkpoint, weight extraction, and polyak updates
all need porting to direct CudaSlice operations.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
10-component design to fix 82% Flat/Small action concentration:
1. Collapse degenerate 9→7 level exposure space
2. Per-branch reward decomposition (direction/magnitude/order/urgency)
3. Hindsight action relabeling for magnitude
4. Position-aware temporal exploration with per-branch epsilon
5. IQN quantile spread as anti-collapse signal
6. Activate PopArt + couple with tau reset
7. Feed the gradient vaccine with diverse batches
8. Per-branch bottleneck widths
9. Ensemble variance at inference
10. Clear PER between folds
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Single train template replaces 7 overlapping workflows. Binary cache
on PVC by commit SHA, fxcache skip-if-exists. Delete 5 dead templates.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Complete bf16 elimination across all crates (ml, ml-core, ml-dqn, ml-ppo,
ml-supervised). Zero half::bf16, __nv_bfloat16, or CudaSlice<half::bf16>
references remain (verified by grep).
CUDA: All 60+ .cu kernels and .cuh headers converted to native float.
- Half-precision intrinsics (__hmul, __hadd, __hdiv) → float operators
- atomicAddBF16 → native atomicAdd(float)
- bf16 wrapper functions → f32 identity passthroughs
Rust: All CudaSlice<half::bf16> → CudaSlice<f32> across 90+ files.
- htod_f32_to_bf16/dtoh_bf16_to_f32 → htod_f32/dtoh_f32 (direct, no conversion)
- Deleted bf16 mirror infrastructure (DuelingWeightSetBf16, alloc_bf16_mirror, etc.)
- Renamed params_bf16→params_flat, d_value_logits_bf16→d_value_logits, etc.
- Fixed .to_f32() sed damage on Decimal::to_f32() and rng.f32()
FxCache: Single f32 disk format (was bf16/f64 dual-version).
- Deleted --bf16 CLI flag from precompute_features
- PVC cache files need regeneration via precompute_features
TF32 tensor cores activated via cublasLtMatmul CUBLAS_COMPUTE_32F — no
explicit TF32 types needed. Storage is pure f32 everywhere.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
CUBLAS_COMPUTE_32F_FAST_TF32 returns NOT_SUPPORTED on RTX 3050 (SM86)
for certain (m,n,k) via cublasLtMatmul (e.g., 128×512×64). Plain
CUBLAS_COMPUTE_32F works for all dimensions. TF32 tensor cores are
still enabled via CUBLAS_TF32_TENSOR_OP_MATH math mode on the cuBLAS
handle — this activates TF32 on Ampere+ automatically when the
hardware supports the specific matrix dimensions.
Also: use sys::cublasLtMatmul directly (bypass result:: wrapper)
for clearer argument ordering.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Root cause of H100 gradient collapse: c51_grad, mse_grad, and cql_grad
kernels wrote d_adv_logits in sample-major interleaved layout [B, TBA]
but cuBLAS backward read it as branch-major [B*B0*NA | B*B1*NA | ...].
At batch=16384, the GEMM read cross-contaminated data from wrong branches,
producing zero weight gradients by symmetry cancellation at scale.
Fix: changed indexing in all 3 grad kernels to branch-major.
Reverted diagnostic hacks (ungraphed forward, forced diagnostics).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- iqn_lambda restored to 0.25 (was 0.0 for NaN isolation)
- STEP_DIAG per-step logging removed (was temporary)
- FOXHUNT_GRAD_DIAG env var removed from Argo template
IQN NaN root cause fixed in 8cbabcef5 (f32-as-bf16 type mismatch).
Ready for H100 baseline deployment.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
C51 cross-entropy structurally prefers low-variance actions (Small positions
have tighter return distributions). MSE is variance-neutral. By reducing
C51's gradient to 20% and amplifying MSE to 4× for the magnitude branch,
MSE becomes the dominant loss signal for position sizing.
Direction/order/urgency keep full C51 gradient for distributional risk awareness.
Replaces the previous mag_amp=1.5 which addressed a symptom (Flat gradient
starvation) rather than the root cause (C51 variance bias).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The mirror universe feature negated only features[0..3] (4 return-like
features) but inverted the direction action (Short↔Long). The remaining
38 directional features (momentum, trend, patterns) were NOT negated.
This taught the model the WRONG direction on 50% of epochs, cancelling
out the directional signal and producing worse-than-random performance.
Disabled until proper full-feature mirroring is implemented.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replaces broken 9-output exposure branch with two 3-output branches.
Each 3-output branch has -1/2 gradient (33%) — strong enough for dueling.
Eliminates i%3 Q-value degeneracy permanently.
3 novel gems: direction-specific dense reward, per-branch epsilon,
Flat detection shortcut with gradient zeroing.
~480 lines of dead code removed (aux optimizer, CEA, pre-training).
20+ files changed. Full CUDA, no CPU path, no stubs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The exposure aux backward GEMM amplifies gradients through the
hidden activation matrix (d_logits × h_b0), producing grad_norm=5000+.
With max_grad_norm=10, this throttles ALL gradients by 500x (lr/500).
Fix: reduce aux_weight 0.5→0.01 (50x) + clip_grad_buf_inplace after
aux GEMM to cap combined gradient at max_grad_norm.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fix 2: Eliminate validation graph re-capture (DtoD into same buffers)
Fix 3: Async validation on separate CUDA stream (overlap with experience)
Fix 4: Aux GEMM in graph_aux (GPU scalar + kernel sig change)
Fix 7: Mega-graph fusion (spectral+forward+aux → 2 launches)
Combined with 5 implemented fixes: 47s → ~16s/epoch
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Pre-training: 50 batch supervised direction init at epoch 0
- Exposure aux targets: DtoD copy from collector to fused context
- PopArt: wired behind config flag (disabled by default)
- TD(λ)/hindsight/curriculum: stubs with config guards
- Loss threshold 500→100K for v8 reward distribution
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
