Journal of Data Science

Black-box machine learning models are recognized as useful tools for prediction applications, but the algorithmic complexity of some models causes interpretation challenges. Explainability methods have been proposed to provide insight into these models, but there is little research focused on supervised modeling with functional data inputs. We argue that, especially in applications of high consequence, it is important to explicitly model the functional dependence in a black-box analysis to not obscure or misrepresent patterns in explanations. As such, we propose the Variable importance Explainable Elastic Shape Analysis (VEESA) pipeline for training supervised machine learning models with functional inputs. The pipeline is an analysis process that includes the data preprocessing, modeling, and post-hoc explanations. The preprocessing is done using elastic functional principal components analysis, which accounts for vertical and horizontal variability in functional data and, ultimately, allows for explanations in the original data space that identify the important functional variability without bias due to correlated variables. Here, we demonstrate the pipeline on two high-consequence applications: explosives classification for national security and inkjet printer identification in forensic science. The applications exhibit the VEESA pipeline’s ability to provide an understanding of the characteristics of the functional data useful for prediction. Code for implementing the pipeline is available in the veesa R package (and supplemental python code).

Attention mechanism has become an almost ubiquitous model architecture in deep learning. One of its distinctive features is to compute non-negative probabilistic distribution to re-weight input representations. This work reconsiders attention weights as bidirectional coefficients instead of probabilistic measures for potential benefits in interpretability and representational capacity. After analyzing the iteration process of attention scores through backwards gradient propagation, we proposed a novel activation function, TanhMax, which possesses several favorable properties to satisfy the requirements of bidirectional attention. We conduct a battery of experiments to validate our analyses and advantages of proposed method on both text and image datasets. The results show that bidirectional attention is effective in revealing input unit’s semantics, presenting more interpretable explanations and increasing the expressive power of attention-based model.

Society’s capacity for algorithmic problem-solving has never been greater. Artificial Intelligence is now applied across more domains than ever, a consequence of powerful abstractions, abundant data, and accessible software. As capabilities have expanded, so have risks, with models often deployed without fully understanding their potential impacts. Interpretable and interactive machine learning aims to make complex models more transparent and controllable, enhancing user agency. This review synthesizes key principles from the growing literature in this field. We first introduce precise vocabulary for discussing interpretability, like the distinction between glass box and explainable models. We then explore connections to classical statistical and design principles, like parsimony and the gulfs of interaction. Basic explainability techniques – including learned embeddings, integrated gradients, and concept bottlenecks – are illustrated with a simple case study. We also review criteria for objectively evaluating interpretability approaches. Throughout, we underscore the importance of considering audience goals when designing interactive data-driven systems. Finally, we outline open challenges and discuss the potential role of data science in addressing them. Code to reproduce all examples can be found at https://go.wisc.edu/3k1ewe.

Abstract: The study of factor analytic models often has to address two im portant issues: (a) the determination of the “optimum” number of factors and (b) the derivation of a unique simple structure whose interpretation is easy and straightforward. The classical approach deals with these two tasks separately, and sometimes resorts to ad-hoc methods. This paper proposes a Bayesian approach to these two important issues, and adapts ideas from stochastic geometry and Bayesian finite mixture modelling to construct an ergodic Markov chain having the posterior distribution of the complete col lection of parameters (including the number of factors) as its equilibrium distribution. The proposed method uses an Automatic Relevance Determi nation (ARD) prior as the device of achieving the desired simple structure. A Gibbs sampler updating scheme is then combined with the simulation of a continuous-time birth-and-death point process to produce a sampling scheme that efficiently explores the posterior distribution of interest. The MCMC sample path obtained from the simulated posterior then provides a flexible ingredient for most of the inferential tasks of interest. Illustrations on both artificial and real tasks are provided, while major difficulties and challenges are discussed, along with ideas for future improvements.