Mean Square Error (MSE), which is commonly used for regression tasks but may not emphasize the magnitude of errors in probabilistic classifications.

Log Loss (Cross-Entropy Loss), which penalizes incorrect classifications more heavily, especially in probabilistic models, making it sensitive to inaccuracies.

Mean Absolute Error (MAE), which averages the absolute differences between predicted and actual values but does not significantly emphasize the magnitude of errors.

Mean Bias Error (MBE), which calculates the average bias in the predictions but is less effective for classification tasks.

Softmax, which is typically used for multi-class classification tasks and not suitable for binary classification.