Views
No views yet
| Task | Algorithm | Metric |
|---|---|---|
| Lap-time prediction | XGBoost | MAE 0.410 s |
| Tire degradation | TCN with Monte Carlo Dropout | P10/P50/P90 quantiles, pit-window detection |
| Overtake probability | LightGBM | AUC-ROC 0.876 |
| Safety-car probability | LightGBM | classifier |
| Pit-stop duration | HistGradientBoosting (quantile) | MAE 0.487 s |
| Undercut success | LightGBM (binary) | AUC-ROC 0.771 |
| Team-radio NLP | Whisper, RoBERTa, SetFit, BERT-large | 4-stage pipeline |
train_seasons: [2023, 2024], test_season: 2025
in each model config). Quoting a figure that pools all three is partly the system reading
back its own training data. The processed data is
published as a companion dataset: VforVitorio/f1-strategy-dataset.1@misc{vegasobral2026f1stratlab,
2 author = {Vega, V{\'i}ctor},
3 title = {F1 StratLab: AI Models for Strategy Recommendations in Formula 1 Races},
4 year = {2026},
5 note = {Bachelor's Thesis, Intelligent Systems Engineering, UIE Campus Coru{\~n}a},
6 url = {https://f1stratlab.com}
7}