About Me

I am a Postdoctoral Researcher at the sAIfer lab, in the Department of Electrical and Electronic Engineering from the University of Cagliari (UNICA), Italy. My research, from a broad perspective, focuses on Adversarial Machine Learning. More in detail, I have specialized on Adversarial Pruning (AP) methods, which aim to reduce the size of a model while preserving/inducing robustness against adversarial attacks.

Short Bio

Giorgio Piras is a Postdoctoral Researcher from the University of Cagliari (UNICA), Italy, where he focuses on AI security. He obtained his BSc (with full marks) in Electrical, Electronic, and Computer Engineering from the University of Cagliari in July 2019. He then obtained, in July 2021, his MSc (with honors) in Computer Engineering; Artificial Intelligence & Cybersecurity from the same university. Finally, he received is PhD (with honors) in the context of the National PhD-AI program, run by La Sapienza University of Rome, Italy, defending the thesis titled “Adversarial Pruning: Improving Evaluations and Methods”, supervised by Prof.Battista Biggio.

During his PhD, he was a visiting student at the Karlsruhe Institute for Technology, Germany, supervised by prof. Christian Wressnegger. He serves constantly as a reviewer for Pattern Recognition Journal and Neurocomputing, and has served at AISEC, ESANN, S+SSPR and ITASEC conferences. He is now involved, with the University of Cagliari, in the EU HORIZON Project “Sec4Ai4Sec”.


Education

  • Postdoctoral Researcher: University of Cagliari, 1st November 2024 - ongoing
  • Ph.D in Artificial Intelligence, University of Rome La Sapienza / University of Cagliari, 24th January 2025 (with honors).
    • Visiting Student: Karlsruhe Institute for Technology (KIT), Karlsruhe, Germany. Supervisor: Prof. Christian Wressnegger.
    • Thesis: Adversarial Pruning: Improving Evaluations and Methods. Supervisor: Prof. Battista Biggio
  • M.S. in Artificial Intelligence & Cybersecurity, University of Cagliari, 27th July 2021 (with honors).
    • Thesis: On Explainability of Machine Learning DGA Detectors from DNS Traffic Data (see related paper). Supervisor: Prof. Battista Biggio
  • B.S. in Electrical, Electronic and Computer Engineering, University of Cagliari, 25th July 2019 (full marks).
    • Thesis: Utilizzabilità di Tecniche Affidabilistiche per la Manutenzione Predittiva di Stazioni Radio Base. Supervisor: Prof. Elio Usai

Projects

  • Participation, with the University of Cagliari, in the EU project ``Security for AI and AI for Security’’ (Sec4AI4Sec), Grant Agreement no.: 101120393, funded by the European Union in the programme HORIZON-CL3-2022-CS-01

Work experience

  • 03/2023 - 06/2023: Newspaper
    • L’Unione Sarda, Cagliari, Italy
    • Magazine articles on AI security for local newspaper.
  • 06/2023 - 07/2023: Consultant
    • Smeralda Computing & Ass. S.r.l., Sassari, Italy
    • Project Report: the state of the Italian cyberSecurity infrastructure.
  • 03/2024 - 06/2024: Newspaper
    • L’Unione Sarda, Cagliari, Italy
    • Magazine articles on AI ethics for local newspaper.
  • 06/2024 - 07/2024: Consultant
    • Smeralda Computing & Ass. S.r.l., Sassari, Italy
    • Project Report: “Kentos” of the Autonomous Region of Sardinia.
  • 12/2024 - 02/2025: AI Security Consultant
    • Consulthink, Rome, Italy
    • Project Report on Secure AI strategies.

Versione Italiana

Sono un Postdoctoral Researcher presso il sAIfer lab, nel Dipartimento di Ingegneria Elettrica ed Elettronica dell’Università di Cagliari (UNICA). La mia ricerca, da una prospettiva generale, si concentra sull’Adversarial Machine Learning. Più nel dettaglio, mi sono specializzato nei metodi di Adversarial Pruning (AP), che mirano a ridurre le dimensioni di un modello di intelligenza artificiale preservandone, o inducendone, la robustezza contro attacchi di tipo adversarial.

Breve Biografia

Giorgio Piras è un Postdoctoral Researcher presso l’Università di Cagliari (UNICA). Ha conseguito la Laurea Triennale in Ingegneria Elettrica, Elettronica e Informatica presso l’Università di Cagliari nel luglio 2019, con voti 110/110. Successivamente, nel luglio 2021, ha conseguito (con lode) la Laurea Magistrale in Computer Engineering; Artificial Intelligence & Cybersecurity presso la medesima università. Infine, ha conseguito il suo dottorato di ricerca (con lode) nel contesto del programma di Dottorato Nazionale in Intelligenza Artificiale (PhD-AI), coordinato dall’Università La Sapienza di Roma, difendendo la tesi dal titolo “Adversarial Pruning: Improving Evaluations and Methods”, supervisionato dal Prof.Battista Biggio. Durante il dottorato, è stato visiting student presso il Karlsruhe Institute for Technology, Germania, sotto la supervisione del prof. Christian Wressnegger. Svolge regolarmente attività di revisore per Pattern Recognition Journal e Neurocomputing, e ha servito come revisore per le conferenze AISEC, ESANN, S+SSPR e ITASEC. Attualmente, mediante l’Università di Cagliari, è coinvolto nel progetto europeo HORIZON “Sec4Ai4Sec”.