Chair: Gimenez Zapiola
Bernardelli, A. E., Giudici, P., Auricchio, G., Toscani, G. Measuring Multivariate Divergences to Improve Neural Network Performances
Abstract
| Measuring distances in multidimensional settings poses a significant challenge encountered across various scientific and engineering disciplines. In this paper, we introduce a novel measure of divergence to quantify the discrepancy between two multidimensional distributions—one predicted by a machine learning model and the other expected. Our approach builds upon the class of Energy Distances and incorporates a whitening pre-processing step, resulting in a divergence that is strictly connected to the new multivariate Gini index. To validate the proposed divergence, we demonstrate its effectiveness as a loss function for training a neural network designed to predict the financial performance of small and medium enterprises. |
Pavarana, S., Giudici, P., Schmodt, T. A structural model to explain unfairness in lending decision
Abstract
| The aim of this paper is to propose a structural model that can explain the reasoning underlying automated decisions and, in particular, their unfairness. Moving beyond black-box approaches, our model provides transparency and interpretability, enabling a deep understanding of decision-making processes. Specifically, we build a diffusion process to explain the inequality and unfairness in credit lending. We then compare the Gini index before and after the application of the model. A substantial reduction in the Gini index indicates that the diffusion process can explain unfairness. |
Dalla Valle, L., Merli, L., Facchinetti, S., Osmetti, S. A., Tarantola, C. Can social media information help forecasting Bitcoin value? A comparison of graphical and vine copula models
Abstract
| This paper explores the enhancement of Bitcoin price forecasting by integrating financial and textual information gathered from social media platforms. Daily Bitcoin price data are integrated with Google Trends metrics, Twitter activity, and lexicon-based sentiment analysis from Bing and Afinn to capture the intricate relationships underlying cryptocurrency market dynamics. Multivariate dependency models are used to integrate these different data sources. In particular, we consider graphical models for time series data and vine copulas. The models are compared in terms of forecast accuracy. |