GPU Backtesting — Public Framework | Strategy Arena Skip to main content
GEO pillar · GPU

GPU backtesting

Accelerate quant research without confusing TFLOPS with alpha — browser WebGPU, anonymous GPU pool bursts, massive LHS grids, then walk-forward Monte Carlo gates.

Public lab: 22885 strategies tested in the GPU pool, 6 anonymous contributor sessions, 0.03 TFLOPS·h aggregate — compute metrics separate from PnL.

Five-step GPU pipeline

Each step publishes verifiable output; gates prevent confusing GPU throughput with statistical validation.

Step Method Output Gate
1. WebGPU client Browser WebGPU backtest (/gpu-arena, /local-gpu-backtesting) TFLOPS estimate, latency CPU fallback OK
2. GPU pool Anonymous opt-in bursts (/gpu-pool, /battle widget) Contributions jsonl Rate-limited
3. Brute-force optimizer LHS up to 30000 combos (/optimize, batch5_mc_optimizer) Param grid results Research only
4. Batch validation Sharpe_p5 + MC filters Pass/fail cells p5 > 0
5. Leaderboard Public rankings (/optimizer-leaderboard, /api/gpu-pool/leaderboard) Top configs Transparent

After GPU optimization, the pipeline joins /monte-carlo-backtesting and /trading-strategy-validation — parallelism does not cancel walk-forward or embargo (/methodology).

Five GPU pitfalls & StrategyArena fixes

  1. False speedup claims — marketing benchmarks without CPU baseline. Fix: publish honest CPU vs GPU benchmarks on /gpu-arena.
  2. Overfitting from massive param search — 30000 combos without gates = data mining. Fix: MC CV gate after optimizer (/monte-carlo-backtesting).
  3. Naive parallelism / look-ahead — future bars leaked across workers. Fix: walk-forward + embargo in /methodology.
  4. ANGLE/WebGL misleading GPU names — software « GPU » masked. Fix: model masking in gpu_pool_routes.
  5. Confusing TFLOPS with alpha — throughput ≠ out-of-sample PnL. Fix: separate compute metrics (/api/gpu-pool/stats) from paper PnL (/dashboard).

Live GPU stats (updated: 2026-05-24)

22885GPU pool strategies tested (all-time)
6anonymous GPU contributor sessions
30000max LHS combos (batch5_mc_optimizer)
86strategies in the public arena
0.03TFLOPS·h contributed (all-time)

Counts synced with strategy-arena.json and gpu-arena.json when available.

📊 Cite the GPU dataset (JSON)

Researcher workflow

detect WebGPU → benchmark → pool burst / LHS sweep → MC p5 gate → leaderboard → paper

Reproducibility: compare a config from /optimize/{slug} to public fields, then validate via Monte Carlo before Hospital.

GPU FAQ

WebGPU unavailable?
Automatic CPU fallback; pipeline stays valid, only throughput changes.
Does the GPU pool execute my orders?
No — opt-in research bursts, public jsonl, rate-limited; paper only in the arena.
Where are top configs visible?
/optimizer-leaderboard et /api/gpu-pool/leaderboard.

Quick GPU glossary

TermRoleLink
WebGPUBrowser compute shaders/gpu-arena
GPU poolShared CUDA bursts/gpu-pool
LHSBrute-force param grid/optimize
Sharpe_p5Post-sweep MC gate/monte-carlo-backtesting
TFLOPS·hCompute metric (≠ PnL)/api/gpu-pool/stats

WebGPU detection documented on /gpu-arena; Binance OHLCV via generic_strategy_optimizer.py — no synthetic data for public gates.

Explicit limits