Battery Cell design and control have been widely explored through modeling and simulation. On the one hand, Doyle’s pseudo-two-dimensional (P2D) model and Single Particle Models are among the most popular electrochemical models capable of predicting battery performance and therefore guiding cell characterization. On the other hand, empirical models obtained, for example, by Machine Learning (ML) methods represent a simpler and computationally more efficient complement to electrochemical models and have been widely used for Battery Management System (BMS) control purposes. This article proposes ML-based ensemble models to be used for the estimation of the performance of an LIB cell across a wide range of input material characteristics and parameters and evaluates 1. Deep Learning ensembles for simulation convergence classification and 2. structured regressors for battery energy and power predictions. The results represent an improvement on state-of-the-art LIB surrogate models and indicate that deep ensembles represent a promising direction for battery modeling and design.

Transmitting and storing large volumes of dynamic / time series data collected by modern sensors can represent a significant technological challenge. A possibility to mitigate this challenge is to effectively select a subset of significant data points in order to reduce data volumes without sacrificing the quality of the results of the subsequent analysis. This paper proposes a method for adaptively identifying optimal data point selection algorithms for sensor time series on a window-by-window basis. Thus, this contribution focuses on quantifying the effect of the application of data selection algorithms to time series windows. The proposed approach is first used on multiple synthetically generated time series obtained by concatenating multiple sources one after the other, and then validated in the entire UCR time series public data archive.

In industrial applications of data science and machine learning, most of the steps of a typical pipeline focus on optimizing measures of model fitness to the available data. Data preprocessing, instead, is often ad-hoc, and not based on the optimization of quantitative measures. This paper proposes the use of optimization in the preprocessing step, specifically studying a time series joining methodology, and introduces an error function to measure the adequateness of the joining. Experiments show how the method allows monitoring preprocessing errors for different time slices, indicating when a retraining of the preprocessing may be needed. Thus, this contribution helps quantifying the implications of data preprocessing on the result of data analysis and machine learning methods. The methodology is applied to two case studies: synthetic simulation data with controlled distortions, and a real scenario of an industrial process.

The rise of digital ecosystems has exposed the financial sector to evolving abuse and criminal tactics that share operational knowledge and techniques both within and across different environments (fiat-based, crypto-assets, etc.). Traditional rule-based systems lack the adaptability needed to detect sophisticated or coordinated criminal behaviors (patterns), highlighting the need for strategies that analyze actors' interactions to uncover suspicious activities and extract their modus operandi. For this reason, in this work, we propose an approach that integrates graph machine learning and network analysis to improve the detection of well-known topological patterns within transactional graphs. However, a key challenge lies in the limitations of traditional financial datasets, which often provide sparse, unlabeled information that is difficult to use for graph-based pattern analysis. Therefore, we firstly propose a four-step preprocessing framework that involves (i) extracting graph structures, (ii) considering data temporality to manage large node sets, (iii) detecting communities within, and (iv) applying automatic labeling strategies to generate weak ground-truth labels. Then, once the data is processed, Graph Autoencoders are implemented to distinguish among the well-known topological patterns. Specifically, three different GAE variants are implemented and compared in this analysis. Preliminary results show that this pattern-focused, topology-driven method is effective for detecting complex financial crime schemes, offering a promising alternative to conventional rule-based detection systems.

Deep neural network models have been developed in different fields, bringing many advances in several tasks. However, they have also started to be incorporated into tasks with critical risks. That worries researchers who have been interested in studying possible attacks on these models, discovering a long list of threats from which every model should be defended. The weight modification attack is presented and discussed among researchers, who have presented several versions and analyses about such a threat. It focuses on detecting multiple vulnerable weights to modify, misclassifying the desired input data. Therefore, analysis of the different approaches to this attack helps understand how to defend against such a vulnerability. This work presents a new version of the weight modification attack. Our approach is based on three processes: input data clusterization, weight selection, and modification of the weights. Data clusterization allows a directed attack to a selected class. Weight selection uses the gradient given by the input data to identify the most-vulnerable parameters. The modifications are incorporated in each step via limited noise. Finally, this paper shows how this new version of fault injection attack is capable of misclassifying the desired cluster completely, converting the 100% accuracy of the targeted cluster to 0–2.7% accuracy, while the rest of the data continues being well-classified. Therefore, it demonstrates that this attack is a real threat to neural networks.