Titulo Estágio
Explainability in Bank Account Fraud Detection using Spiking Neural Networks
Áreas de especialidade
Sistemas Inteligentes
Local do Estágio
CISUC
Enquadramento
The advancements in Deep Neural Networks (DNNs) have led to the development of powerful and intricate solutions, but their black-box nature often fails to provide explanations for their decision-making process. This lack of transparency undermines the trustworthinesst in these systems, thereby restricting their applicability and hindering their full potential, particularly when they are integrated into sensitive decision-making processes.
In contrast, Spiking Neural Networks (SNNs) closely emulate biological neurons and possess characteristics like sparsity and energy efficiency, providing distinct advantages compared to DNNs. Leveraging these models, especially in critical applications such as security/safety systems, health diagnosis, and fraud detection, necessitates addressing the challenges of trustworthiness and explainability. It is essential to ensure that end-users are convinced of the reliability of SNNs and have confidence in their outputs.
In this internship it is proposed to research and implement explainability mechanisms within spiking learning models, specifically in the context of financial applications. The goal is to develop a transparent and explainable solution for responsible-based financial models in bank account AI. The internship is part of the NextGenAI - Center for responsible AI - Project, which is dedicated to finding solutions in the field of responsible artificial intelligence.
Objetivo
In this internship the student should study, propose, implement, and test methods for explainability and trustworthiness in spiking neural networks
To achieve this goal, the following objectives will be pursued:
- Study the state of the art
- Study the available frameworks for model development
- Define case studies for fraud financial applications
- Define, implement, and fine tune the explainable spiking neural network architecture
- Propose and deploy test setup
Plano de Trabalhos - Semestre 1
- Literature review
- Identification and study of explainable mechanisms
- Identification and study of available tools and frameworks
- Analyse and define case studies
- Define the architecture of the system
- Start implementing the proposed approach
- Write intermediate report
Plano de Trabalhos - Semestre 2
- Implement the proposed solution and fine tune models
- Test and evaluate performance
- Write final report
Condições
The internship will be held at the Laboratory of Artificial Neural Networks (LARN-CISUC), offering an opportunity to collaborate with a diverse network of academic institutions, companies, and end-users through an AI Consortium. The selected student should possess a strong motivation to conduct research in the fields of artificial intelligence, machine learning, and data science. Additionally, there is a potential research grant opportunity of XXX Euros available under the research program of the NextGenAI - Centre For Responsible AI project, supported by the National Recovery and Resilience Plan (PRR). The availability of the grant will be contingent upon the student's performance during the internship.
Orientador
Bernardete Ribeiro e Catarina Silva
catarina@dei.uc.pt 📩