[ [ Reviewer 1 ] ]

Thank you for your feedback. The loss function should be selected based on the application, e.g., if We will add a discussion on loss functions and regularization in the final manuscript. Thank you for your helpful comments and suggestions. As mentioned in lines 75 and 89, different functional forms are deemed interpretable in different applications. Reviewer 3.) The theoretical justification of our framework was provided in Section 3.1, where we have shown that Our algorithm explores the Pareto front of simplicity vs. predictivity systematically We will add all the suggested references in the final the manuscript.