Tree ensembles achieve state-of-the-art performance on numerous prediction tasks. We propose Joint Optimization of Piecewise Linear ENsembles (JOPLEN), which jointly fits piecewise linear models at all leaf nodes of an existing tree ensemble. In addition to enhancing the expressiveness of an ensemble, JOPLEN allows several common penalties, including sparsity-promoting matrix norms and subspace-norms, to be applied to nonlinear prediction. We demonstrate the performance of JOPLEN on over 100 regression and classification datasets and with a variety of penalties. JOPLEN leads to improved prediction performance relative to not only standard random forest and gradient boosted tree ensembles, but also other methods for enhancing tree ensembles. We demonstrate that JOPLEN with a nuclear norm penalty learns subspace-aligned functions. Additionally, JOPLEN combined with a Dirty LASSO penalty is an effective feature selection method for nonlinear prediction in multitask learning.

The rise of machine learning methods on heavily resource constrained devices requires not only the choice of a suitable model architecture for the target platform, but also the optimization of the chosen model with regard to execution time consumption for inference in order to optimally utilize the available resources. Random forests and decision trees are shown to be a suitable model for such a scenario, since they are not only heavily tunable towards the total model size, but also offer a high potential for optimizing their executions according to the underlying memory architecture. In addition to the straightforward strategy of enforcing shorter paths through decision trees and hence reducing the execution time for inference, hardware-aware implementations can optimize the execution time in an orthogonal manner. One particular hardware-aware optimization is to layout the memory of decision trees in such a way, that higher probably paths are less likely to be evicted from system caches. This works particularly well when splits within tree nodes are uneven and have a high probability to visit one of the child nodes. In this paper, we present a method to reduce path lengths by rewarding uneven probability distributions during the training of decision trees at the cost of a minimal accuracy degradation. Specifically, we regularize the impurity computation of the CART algorithm in order to favor not only low impurity, but also highly asymmetric distributions for the evaluation of split criteria and hence offer a high optimization potential for a memory architecture-aware implementation. We show that especially for binary classification data sets and data sets with many samples, this form of regularization can lead to an reduction of up to approximately four times in the execution time with a minimal accuracy degradation.

Computational notebooks have become the primary coding environment for data scientists. However, research on their code quality is still emerging, and the code shared is often of poor quality. Given the importance of maintenance and reusability, understanding the metrics that affect notebook code comprehensibility is crucial. Code understandability, a qualitative variable, is closely tied to user opinions. Traditional approaches to measuring it either use limited questionnaires to review a few code pieces or rely on metadata such as likes and votes in software repositories. Our approach enhances the measurement of Jupyter notebook understandability by leveraging user comments related to code understandability. As a case study, we used 542,051 Kaggle Jupyter notebooks from our previous research, named DistilKaggle. We employed a fine-tuned DistilBERT transformer to identify user comments associated with code understandability. We established a criterion called User Opinion Code Understandability (UOCU), which considers the number of relevant comments, upvotes on those comments, total notebook views, and total notebook upvotes. UOCU proved to be more effective than previous methods. Furthermore, we trained machine learning models to predict notebook code understandability based solely on their metrics. We collected 34 metrics for 132,723 final notebooks as features in our dataset, using UOCU as the label. Our predictive model, using the Random Forest classifier, achieved 89% accuracy in predicting the understandability levels of computational notebooks.

Explainable machine learning (XML) has emerged as a major challenge in artificial intelligence (AI). Although black-box models such as Deep Neural Networks and Gradient Boosting often exhibit exceptional predictive accuracy, their lack of interpretability is a notable drawback, particularly in domains requiring transparency and trust. This paper tackles this core AI problem by proposing a novel method to enhance explainability with minimal accuracy loss, using a Mixture of Linear Models (MLM) estimated under the co-supervision of black-box models. We have developed novel methods for estimating MLM by leveraging AI techniques. Specifically, we explore two approaches for partitioning the input space: agglomerative clustering and decision trees. The agglomerative clustering approach provides greater flexibility in model construction, while the decision tree approach further enhances explainability, yielding a decision tree model with linear or logistic regression models at its leaf nodes. Comparative analyses with widely-used and state-of-the-art predictive models demonstrate the effectiveness of our proposed methods. Experimental results show that statistical models can significantly enhance the explainability of AI, thereby broadening their potential for real-world applications. Our findings highlight the critical role that statistical methodologies can play in advancing explainable AI.

Can a deep neural network be approximated by a small decision tree based on simple features? This question and its variants are behind the growing demand for machine learning models that are *interpretable* by humans. In this work we study such questions by introducing *interpretable approximations*, a notion that captures the idea of approximating a target concept $c$ by a small aggregation of concepts from some base class $\mathcal{H}$. In particular, we consider the approximation of a binary concept $c$ by decision trees based on a simple class $\mathcal{H}$ (e.g., of bounded VC dimension), and use the tree depth as a measure of complexity. Our primary contribution is the following remarkable trichotomy. For any given pair of $\mathcal{H}$ and $c$, exactly one of these cases holds: (i) $c$ cannot be approximated by $\mathcal{H}$ with arbitrary accuracy; (ii) $c$ can be approximated by $\mathcal{H}$ with arbitrary accuracy, but there exists no universal rate that bounds the complexity of the approximations as a function of the accuracy; or (iii) there exists a constant $\kappa$ that depends only on $\mathcal{H}$ and $c$ such that, for *any* data distribution and *any* desired accuracy level, $c$ can be approximated by $\mathcal{H}$ with a complexity not exceeding $\kappa$. This taxonomy stands in stark contrast to the landscape of supervised classification, which offers a complex array of distribution-free and universally learnable scenarios. We show that, in the case of interpretable approximations, even a slightly nontrivial a-priori guarantee on the complexity of approximations implies approximations with constant (distribution-free and accuracy-free) complexity. We extend our trichotomy to classes $\mathcal{H}$ of unbounded VC dimension and give characterizations of interpretability based on the algebra generated by $\mathcal{H}$.

Fine-resolution spatial patterns of water table depth (WTD) play a crucial role in shaping ecological resilience, hydrological connectivity, and anthropocentric objectives. Generally, a large-scale (e.g., continental or global) spatial map of static WTD can be simulated using either physically-based (PB) or machine learning-based (ML) models. We construct three fine-resolution (500 m) ML simulations of WTD, using the XGBoost algorithm and more than 20 million real and proxy observations of WTD, across the United States and Canada. The three ML models were constrained using known physical relations between WTD's drivers and WTD and were trained by sequentially adding real and proxy observations of WTD. We interpret the black box of our physically constrained ML models and compare it against available literature in groundwater hydrology. Through an extensive (pixel-by-pixel) evaluation, we demonstrate that our models can more accurately predict unseen real and proxy observations of WTD across most of North America's ecoregions compared to three available PB simulations of WTD. However, we still argue that large-scale WTD estimation is far from being a solved problem. We reason that due to biased observational data mainly collected from low-elevation floodplains, the misspecification of equations within physically-based models, and the over-flexibility of machine learning models, verifiably accurate simulations of WTD do not yet exist. Ultimately, we thoroughly discuss future directions that may help hydrogeologists decide how to proceed with WTD estimations, with a particular focus on the application of machine learning and the use of proxy satellite data.

We extend decision tree and random forest algorithms to mixed-curvature product spaces. Such spaces, defined as Cartesian products of Euclidean, hyperspherical, and hyperbolic manifolds, can often embed points from pairwise distances with much lower distortion than in single manifolds. To date, all classifiers for product spaces fit a single linear decision boundary, and no regressor has been described. Our method overcomes these limitations by enabling simple, expressive classification and regression in product manifolds. We demonstrate the superior accuracy of our tool compared to Euclidean methods operating in the ambient space for component manifolds covering a wide range of curvatures, as well as on a selection of product manifolds.

The value of supervised dimensionality reduction lies in its ability to uncover meaningful connections between data features and labels. Common dimensionality reduction methods embed a set of fixed, latent points, but are not capable of generalizing to an unseen test set. In this paper, we provide an out-of-sample extension method for the random forest-based supervised dimensionality reduction method, RF-PHATE, combining information learned from the random forest model with the function-learning capabilities of autoencoders. Through quantitative assessment of various autoencoder architectures, we identify that networks that reconstruct random forest proximities are more robust for the embedding extension problem. Furthermore, by leveraging proximity-based prototypes, we achieve a 40% reduction in training time without compromising extension quality. Our method does not require label information for out-of-sample points, thus serving as a semi-supervised method, and can achieve consistent quality using only 10% of the training data.

Accurate determination of the geothermal gradient is critical for assessing the geothermal energy potential of a given region. Of particular interest is the case of Colombia, a country with abundant geothermal resources. A history of active oil and gas exploration and production has left drilled boreholes in different geological settings, providing direct measurements of the geothermal gradient. Unfortunately, large regions of the country where geothermal resources might exist lack such measurements. Indirect geophysical measurements are costly and difficult to perform at regional scales. Computational thermal models could be constructed, but they require very detailed knowledge of the underlying geology and uniform sampling of subsurface temperatures to be well-constrained. We present an alternative approach that leverages recent advances in supervised machine learning and available direct measurements to predict the geothermal gradient in regions where only global-scale geophysical datasets and course geological knowledge are available. We find that a Gradient Boosted Regression Tree algorithm yields optimal predictions and extensively validate the trained model. We show that predictions of our model are within 12% accuracy and that independent measurements performed by other authors agree well with our model. Finnally, we present a geothermal gradient map for Colombia that highlights regions where futher exploration and data collection should be performed.

We propose an innovative statistical method, called Ordinal Mixed-Effect Random Forest (OMERF), that extends the use of random forest to the analysis of hierarchical data and ordinal responses. The model preserves the flexibility and ability of modeling complex patterns of both categorical and continuous variables, typical of tree-based ensemble methods, and, at the same time, takes into account the structure of hierarchical data, modeling the dependence structure induced by the grouping and allowing statistical inference at all data levels. A simulation study is conducted to validate the performance of the proposed method and to compare it to the one of other state-of-the art models. The application of OMERF is exemplified in a case study focusing on predicting students performances using data from the Programme for International Student Assessment (PISA) 2022. The model identifies discriminating student characteristics and estimates the school-effect.