| Issue | #Downvotes for this reason | By |
|---|
| A | The AIC is used to choose the best iteration. | |
| B | Binary splits (convert nominal attributes to binary ones) | |
| C | Use cross-validation for boosting at all nodes (i.e., disable heuristic) | |
| I | Set fixed number of iterations for LogitBoost (instead of using cross-validation) | default: -1 |
| M | Set minimum number of instances at which a node can be split (default 15) | default: 15 |
| P | Use error on probabilities instead of misclassification error for stopping criterion of LogitBoost. | |
| R | Split on residuals instead of class values | |
| W | Set beta for weight trimming for LogitBoost. Set to 0 (default) for no weight trimming. | default: 0.0 |