Machine-learning solutions for pattern classification problems are nowadays widely deployed in society and industry. However, the lack of transparency and accountability of most accurate models often hinders their safe use. Thus, there is a clear need for developing explainable artificial intelligence mechanisms. There exist model-agnostic methods that summarize feature contributions, but their interpretability is limited to predictions made by black-box models. An open challenge is to develop models that have intrinsic interpretability and produce their own explanations, even for classes of models that are traditionally considered black boxes like (recurrent) neural networks. In this article, we propose a long-term cognitive network (LTCN) for interpretable pattern classification of structured data. Our method brings its own mechanism for providing explanations by quantifying the relevance of each feature in the decision process. For supporting the interpretability without affecting the performance, the model incorporates more flexibility through a quasi-nonlinear reasoning rule that allows controlling nonlinearity. Besides, we propose a recurrence-aware decision model that evades the issues posed by the unique fixed point while introducing a deterministic learning algorithm to compute the tunable parameters. The simulations show that our interpretable model obtains competitive results when compared to state-of-the-art white and black-box models.

Traditionally, our understanding of how proteins operate and how evolution shapes them is based on two main data sources: the overall protein fold and the protein amino acid sequence. However, a significant part of the proteome shows highly dynamic and/or structurally ambiguous behavior, which cannot be correctly represented by the traditional fixed set of static coordinates. Representing such protein behaviors remains challenging and necessarily involves a complex interpretation of conformational states, including probabilistic descriptions. Relating protein dynamics and multiple conformations to their function as well as their physiological context (e.g., post-translational modifications and subcellular localization), therefore, remains elusive for much of the proteome, with studies to investigate the effect of protein dynamics relying heavily on computational models. We here investigate the possibility of delineating three classes of protein conformational behavior: order, disorder, and ambiguity. These definitions are explored based on three different datasets, using interpretable machine learning from a set of features, from AlphaFold2 to sequence-based predictions, to understand the overlap and differences between these datasets. This forms the basis for a discussion on the current limitations in describing the behavior of dynamic and ambiguous proteins.

Constricted Particle Swarm Optimization based Algorithm for Global Optimization

Fuzzy-rough cognitive networks (FRCNs) are recurrent neural networks (RNNs) intended for structured classification purposes in which the problem is described by an explicit set of features. The advantage of this granular neural system relies on its transparency and simplicity while being competitive to state-of-the-art classifiers. Despite their relative empirical success in terms of prediction rates, there are limited studies on FRCNs' dynamic properties and how their building blocks contribute to the algorithm's performance. In this article, we theoretically study these issues and conclude that boundary and negative neurons always converge to a unique fixed-point attractor. Moreover, we demonstrate that negative neurons have no impact on the algorithm's performance and that the ranking of positive neurons is invariant. Moved by our theoretical findings, we propose two simpler fuzzy-rough classifiers that overcome the detected issues and maintain the competitive prediction rates of this classifier. Toward the end, we present a case study concerned with image classification, in which a convolutional neural network is coupled with one of the simpler models derived from the theoretical analysis of the FRCN model. The numerical simulations suggest that once the features have been extracted, our granular neural system performs as well as other RNNs.