PhD on Multi-criteria,-modal,-score&-optimization Fin. Forensics&Analytics

JADS is seeking enthusiastic candidates for the position of PhD student in Multi-criteria, -modal, -score and -optimization Financial Forensics & Analytics. On the one hand, multi-criteria decision making is a family of approaches to decision-making and group-oriented negotiation and risk management processes part of combinatorial optimization. On the other hand, financial transactions—from small scale micro-audits to larger-scale cryptomining—require multi-grain inception, representation and reasoning approaches to cope with the sizes, shapes, scale, and volume of transactions to be handled. The candidate shall operationalize this huge ambition by combining the state of the art in hybrid AI with frontier financial analytics, risk modelling and more. Finally, the candidate shall strive to explicitly align our research agenda on AI with the United Nation’s sustainable development goals. The project is funded by Deloitte (The Netherlands).

Short Description

Given its role in modern financial systems, financial distress is a major phenomenon worthy of considerable interest from several perspectives, especially in the scope of financial informatics and predictive modeling/forecasting of the phenomenon’s predictors as well as its circumstantial scientific ramifications such as automated Fintech Computing or Financial Fatigue assessment.

To aid in the modelling of financial distress beyond traditional ratio analysis, scores such as Altman’s Z are often used as a reference, but they tend to require inaccessible data or complex predictive scoring exercises by the hand of expensive human analysts. In addition, and probably more problematically, currently applications of perceptron/deep neural networks are typically a-symmetric and tightly coupled with a particular at hand. In particular, they focus on the modelling and analytics exercise only, while neglecting the underlying data pipeline and its automated governance[1] as well as the data productization and discovery opportunities, which remain to this day largely unexplored. That makes such solutions as rather brittle, failing to meet the expectations of business, and exceedingly case-specific in nature. This makes actionable research to resolve these challenges of critical importance.

Are you an enthusiastic and ambitious researcher with a completed master’s degree in a field related to machine learning (Computer science, AI, Data Science) or in Econometrics with an affinity for AI and deep learning? Does the idea of working on real-world problems and with Deloitte excite you? And are you passionate about using trustworthy hybrid AI methods for the next generation of actionable methods and tools that foster a multi-score, multi-criteria, multi-optimization cube capable of offering a multi-modal and explainable overview (M-Cube) over credit risk modelling, scoring, and asset management?

We are actively recruiting a Ph.D. candidate who will develop and validate novel concepts, methods, and tools for M-Cube capable of delivering a trustworthy overview and understanding over credit risk modelling, scoring, and asset management, and trial them with industrial partners who work with Deloitte.

[1] Desheng Wu, Xiyuan Ma, David L. Olson,, Financial distress prediction using integrated Z-score and multilayer perceptron neural networks,Decision Support Systems, Volume 159, 2022.

Job Description

This vacancy falls under the auspices of the JADE lab, which is the data/AI engineering and governance research UNIT of the JADS, and DELOITTE. In particular, this position will be aligned within the governance of the existing Deloitte lab on Auditing for Responsible AI Software System. This means for example that you will work in a team of 5 other PhD students.

The industrial setting of the deep involvement of Deloitte will balance the rigour with relevance and ascertain fit with societal requirements and trends, validation with industrial case studies.

Job description

Please find the full job desciption on the website of TU/e

Information and application

Application can only be done through the online portal of TU/e (see button below). Applications via regular email will not be taken into consideration.