Searcharxiv⌕ Search

arXiv subjects

Cristina Molero-Río

Publications and source records attributed to Cristina Molero-Río.

4 recordsLinked to original sources

A conic optimization framework for shape-constrained functional regression analysis

Standard regression methods do not always guarantee that the estimated model reflects prior knowledge about how predictors relate to the response, which motivates constrained regression. When the effects to be estimated are smooth nonlinear functions, a natural family of constraints is that of shape constraints, such as sign, monotonicity, or curvature, which encode qualitative properties that these functions are often expected to satisfy. While shape-constrained estimation of smooth functions is well established in classical regression with scalar covariates and response, its extension to functional regression, where the coefficients are themselves functions, remains largely unexplored. In this paper, we propose the functional shape-constrained conic P-spline estimator (FunSCOPE), a unified framework for shape-constrained functional linear regression that covers scalar-on-function, function-on-scalar, and function-on-function models, and in which any combination of the shapes above can be imposed on the functional regression coefficient. Building on a P-spline approach, the desired shapes are expressed as conditions on the scalar regression coefficients, which leads to an estimation problem formulated as a conic optimization problem. For univariate coefficients, as in scalar-on-function, function-on-scalar, and a particular case of the function-on-function regression, the concurrent case, these conditions are both necessary and sufficient, in contrast to existing approaches, which provide only sufficient conditions; for bivariate coefficients, as in general function-on-function models, they are neither necessary nor sufficient. Numerical experiments show that FunSCOPE outperforms competing methods in terms of estimation error and computational time.

stat.ME↗

Optimal Risk Scores for Continuous Predictors

In this paper, we propose a novel Mixed-Integer Non-Linear Optimization formulation to construct a risk score, where we optimize the logistic loss with sparsity constraints. Previous approaches are typically designed to handle binary datasets, where continuous predictor variables are discretized in a preprocessing step by using arbitrary thresholds, such as quantiles. In contrast, we allow the model to decide for each continuous predictor variable the particular threshold that is critical for prediction. The usefulness of the resulting optimization problem is tested in synthetic datasets.

math.OC↗

Optimal randomized classification trees

Classification and Regression Trees (CARTs) are off-the-shelf techniques in modern Statistics and Machine Learning. CARTs are traditionally built by means of a greedy procedure, sequentially deciding the splitting predictor variable(s) and the associated threshold. This greedy approach trains trees very fast, but, by its nature, their classification accuracy may not be competitive against other state-of-the-art procedures. Moreover, controlling critical issues, such as the misclassification rates in each of the classes, is difficult. To address these shortcomings, optimal decision trees have been recently proposed in the literature, which use discrete decision variables to model the path each observation will follow in the tree. Instead, we propose a new approach based on continuous optimization. Our classifier can be seen as a randomized tree, since at each node of the decision tree a random decision is made. The computational experience reported demonstrates the good performance of our procedure.

stat.ML↗

Sparsity in Optimal Randomized Classification Trees

Decision trees are popular Classification and Regression tools and, when small-sized, easy to interpret. Traditionally, a greedy approach has been used to build the trees, yielding a very fast training process; however, controlling sparsity (a proxy for interpretability) is challenging. In recent studies, optimal decision trees, where all decisions are optimized simultaneously, have shown a better learning performance, especially when oblique cuts are implemented. In this paper, we propose a continuous optimization approach to build sparse optimal classification trees, based on oblique cuts, with the aim of using fewer predictor variables in the cuts as well as along the whole tree. Both types of sparsity, namely local and global, are modeled by means of regularizations with polyhedral norms. The computational experience reported supports the usefulness of our methodology. In all our data sets, local and global sparsity can be improved without harming classification accuracy. Unlike greedy approaches, our ability to easily trade in some of our classification accuracy for a gain in global sparsity is shown.

math.OC↗