| KEYWORD |
Implementation of 'fairness by design' principles and fairness metrics in AI systems
Thesis in external company
keywords ALGORITHM FAIRNESS, ARTIFICIAL INTELLIGENCE, DATA ETHICS, DATA QUALITY, DATA SCIENCE, EXPLAINABLE AI, HUMAN-COMPUTER INTERACTION, SOFTWARE ENGINEERING, SYNTHETIC DATA
Reference persons MARCO TORCHIANO, ANTONIO VETRO'
Research Groups DAUIN - GR-16 - SOFTWARE ENGINEERING GROUP - SOFTENG, DAUIN - GR-22 - Nexa Center for Internet & Society - NEXA
Thesis type EXPEIRMENTAL, IN COMPANY, INDUSTRIAL, RESEARCH / EXPERIMENTAL, RESEARCH THESIS WITH A COMPANY, RESEARCH, INNOVATIVE
Description This thesis topic is critically important in the rapidly evolving realm of AI regulation, emphasizing the integration of AI Act stipulations into technological processes. It specifically targets the implementation of 'fairness by design' principles and fairness metrics in AI systems to uphold both regulatory compliance while testing technical feasibility.
The thesis will elaborate on a systematic method during the data preprocessing phase aimed at ensuring diversity and representativeness in training datasets, including using data augmentation techniques. Expected processes include automated bias detection and correction mechanisms within these workflows, as well as benchmarking these processes against the technical conformity assessments mandated by the AI Act.
**
Clearbox AI is an innovative SME, incubated in I3P, winner of the National Innovation Award (PNI 2019) in the ICT category and the EU Seal of Excellence awarded by the European Commission. Clearbox AI is developing a unique and innovative technology ("AI Control Room"), which allows to put into production artificial intelligence models that are robust, explainable and monitorable over time.
See also https://www.clearbox.ai
Required skills Good programming skills and basic knowledge of common data analytics tools and techniques. Grade point average equal to or higher than 26 will play a relevant role in the selection.
Notes When sending your application, we kindly ask you to attach the following information:
- list of exams taken in you master degree, with grades and grade point average
- a résumé or equivalent (e.g., linkedin profile), if you already have one
- by when you aim to graduate and an estimate of the time you can devote to the thesis in a typical week
Deadline 15/03/2025
PROPONI LA TUA CANDIDATURA