Andrea Mazzali

PhD in Artificial Intelligence

Intelligent systems,for Finance.

Exploring the intersection of artificial intelligence, statistics and financial systems.

Scroll to explore

01 · About

Research at the intersection of AI, statistics and finance.

I explore how artificial intelligence and statistical methods can be used to understand complex financial systems and develop more intelligent, reliable and data-driven decision tools.

My research focuses on machine learning, synthetic data generation and quantitative modelling, with particular attention to banking and financial applications. I am interested in building systems that combine methodological rigour with practical relevance.

Portrait of Andrea Mazzali
Andrea MazzaliPhD Student in Artificial Intelligence
Currently
PhD Student in Artificial Intelligence
Institution
University of Bergamo
Research focus
Artificial Intelligence · Statistics · Finance
Academic background
Master's Degree in FinanceUniversity of Pavia
Bachelor's Degree in EconomicsUniversity of Pavia
Artificial intelligenceStatistical modellingFinancial systems

02 · Research

Research directions shaped by data, intelligence and finance.

01

Synthetic data generation for banking applications

Developing statistical and deep learning methods to generate realistic synthetic financial data while preserving structure, utility and confidentiality.

Synthetic dataBankingGenerative modelsPrivacy
02

AI-driven exit strategy modelling in private equity

Using company fundamentals, market signals and predictive models to support decisions on exit timing, route selection and valuation under changing financial conditions.

Private equityExit timingIPOTrade saleValuation
03

Statistical and machine learning methods for finance

Studying how probabilistic modelling, machine learning and network-based methods can improve inference and decision-making in complex financial systems.

Machine learningStatisticsNetworksFinancial systems

Research network

Explore the relationships behind my research.

Select a node to explore how methods, applications and financial research areas are connected.

Machine learning enables Artificial intelligenceDeep learning extends Artificial intelligenceStatistics supports Artificial intelligenceArtificial intelligence enables Synthetic dataArtificial intelligence applies to Exit strategiesMachine learning applies to Financial systemsMachine learning applies to Private equityDeep learning enables Synthetic dataSynthetic data applies to BankingStatistics supports Synthetic dataStatistics supports Network methodsStatistics applies to BankingStatistics applies to Financial systemsNetwork methods applies to BankingNetwork methods informs Private equityPrivate equity informs Exit strategiesFinancial systems informs Private equityFinancial systems informs Exit strategiesArtificial intelligenceMachine learningDeep learningSynthetic dataBankingStatisticsNetwork methodsPrivate equityExit strategiesFinancial systems

AI research assistant

Ask questions about my research.

The assistant interprets your question in relation to the research entity selected in the network.

0 / 2000

03 · Publications

Selected research, automatically connected.

Publications are analysed automatically, assigned to the relevant research areas and connected to the interactive knowledge network.

01
2026Journal of the Royal Statistical Society Series A: Statistics in Society

A network-based distributional inference approach to model country-level cyber risk

Alessandro Spelta, Silvia Facchinetti, Silvia Angela Osmetti, et al.

This article leverages Wasserstein Propagation in Social Network to propose a novel distributional framework for the inference of cyber risk across interconnected economic systems. Cyber attacks represent an increasing threat to global security and economic stability, making the assessment of cyber risk…

Network methodsStatistics
Read publication

04 · Articles

Research ideas, explained beyond the paper.

Articles explore research methods, applications and ideas through a more accessible format, while remaining connected to the underlying publications, formulas and research network.

Explore all articles
01
5 August 202635 min read33 equations

Mapping Transactional Behaviour in Banking Networks: A Wasserstein Propagation Approach to Account-Level Financial Profiles

Andrea Mazzali

Banking transaction data contain rich information on financial behaviour, customer interactions and potential risk patterns. However, their sensitivity, heterogeneity and relational structure make them difficult to analyse, share and reconstruct in a statistically meaningful way. This paper proposes a network-based…

NetworksBankingFinancial systemsStatistics
Read article