0.4404
Avg p-adic loss
2,407
Avg non-zero coefficients
13,813
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 | 0.4561 | 2,405 | 13,813 | 1,191 |
| Fold 1 | 0.4095 | 2,398 | 13,813 | 1,230 |
| Fold 2 | 0.4336 | 2,418 | 13,813 | 1,237 |
| Fold 3 | 0.4580 | 2,405 | 13,813 | 1,211 |
| Fold 4 | 0.4451 | 2,407 | 13,813 | 1,239 |