# Эксперимент: Bounded predictive-gain optimizer (#768)

{
  "worked": true,
  "confidence": 9,
  "verdict": "Built a bounded predictive-gain optimizer with Euclidean mirror descent, groupwise delayed direction products, clipping, persistence, and a deterministic mini-training comparison. All three mechanism predictions were confirmed: gain response slope 0.08450000000000002 predicted versus 0.08450000000000005 observed, persistence ratio 0.6666666666666666 exactly matched, and strict quadratic sign reversal began on the first grid point above the predicted aλ=1 threshold while the stability interval test had zero mismatches. The toy baseline and adaptive optimizer had effectively identical final accuracy (0.93375), with final losses 0.1388764 and 0.1388563 respectively, so no meaningful optimization win was observed.",
  "metrics": {
    "baseline": "final_loss=0.1388764, best_loss=0.0571861, final_accuracy=0.93375, loss_at_40=0.1623599",
    "idea": "final_loss=0.1388563, best_loss=0.0571893, final_accuracy=0.93375, loss_at_40=0.1624318, final_gains=[0.7997841,0.7996852], reversal_events=185, clip_events=0"
  },
  "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py",
  "files": [
    "experiment.py",
    "results.json"
  ],
  "limitations": "The experiment used a small deterministic NumPy MLP and synthetic 2D data, not CIFAR-10, a CNN, Transformer, AdamW, GPU execution, or equal-FLOP profiling. It did not test deliberate learning-rate miscalibration, distribution shift, scalar-per-parameter gains, or statistically replicated seeds."
}