0.4148
Avg p-adic loss
2,208
Avg non-zero coefficients
10,000
Avg iterations
6,108
Total test samples
Method Description
This model implements Zubarev's p-adic polynomial regression method from arXiv:2503.23488. Key features:
- Initialization: Starts from all zeros
- Optimization: Simulated annealing with temperature-controlled random walk
- Perturbations: P-adic-aware (prefers powers of 71)
- Optional: Mahler polynomial basis for non-linear transforms
Cross-validation Results
| Fold | Mean p-adic loss | Non-zero coeffs | Iterations | Test samples |
|---|---|---|---|---|
| Fold 0 | 0.4176 | 2,198 | 10,000 | 1,191 |
| Fold 1 | 0.3808 | 2,220 | 10,000 | 1,230 |
| Fold 2 | 0.4223 | 2,212 | 10,000 | 1,237 |
| Fold 3 | 0.4146 | 2,210 | 10,000 | 1,211 |
| Fold 4 | 0.4386 | 2,198 | 10,000 | 1,239 |