Hi! My name is Flavia Salutari, I was born in L’Aquila, a lovely town in Central Italy, on November 30th 1993.
Until June 2021 I was a Computer Science PhD student at Telecom Paris and Institut Polytechnique de Paris, under the supervision of Mauro Sozio and Dario Rossi.
Before, I was a visiting PhD student at the University College London (UCL) in the Web Intelligence Group and in the SpaceTimeLab working under the supervision of Aldo Lipani on the measurement of societal biases embedded in language models.
Previously, I worked 9 months as Research and Development engineer at Telecom Paris, under the supervision of Dario Rossi.
I obtained both my B.Sc. (TLC Engineering July’15) and my M.Sc. (ICT for Smart Societies October’17) at Politecnico di Torino.
More details can be found by reading my cv (that can as well be downloaded in its more compact version as a pdf).
I have a strong intersciplinary background, and as such I am interested in the applications of Computer and Data Science knowledge in several domains, in particular those who are sustainable and positive for the society.
Recently, I have been working on ethical AI. Specifically, I focused on the measurement and analysis of linguistic bias embedded in transformer-based language models, e.g. BERT, RoBERTa, GPT-2, etc.
During my PhD, I had the privilege to work with human-labeled data coming from real browsing measurements thanks to a collaboration with the Wikimedia Foundation aimed at studying the users’ performance perception on Wikipedia pages. In this context, I worked on the analysis and on the prediction of Quality of Experience (QoE) of Web users, a topic which covers various disciplines, notably computer networks, social psychology, cognitive science and economics.
You can find my publications and technical reports in this section of the website or by visiting my Google Scholar.
Why did I feel this urge to start writing a blog? The answer lies probably in the fact that blogging is a common need for academics. In the blog section of this site you may find posts related to different topics.