Skip to content
View Mic52M's full-sized avatar

Block or report Mic52M

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Mic52M/README.md

Michele Mastroberti

Cybersecurity Researcher | AI Security & ML Specialist | University of Milan

I'm a cybersecurity researcher specializing in AI security and machine learning safety, currently conducting research on assurance evaluation methodologies for artificial intelligence models. My focus is on robustness, fairness, and compliance of AI systems.

🔬 Current Research

Assurance Evaluation of AI Models - Master's Research Project
Defining and implementing assurance checks for AI models, focusing on compliance with the European AI Act. The research evaluates non-functional properties such as robustness against attacks and model fairness.

AI Security & Compliance Analysis
Developing frameworks for AI system security analysis, including vulnerability assessments and testing methodologies for supply chain security.

💻 Tech Stack

Python Bash Kubernetes Docker Go

📚 Academic Background

  • M.Sc. Cybersecurity - University of Milan (2023-2025)
  • B.Sc. Systems & Network Security - University of Milan (2020-2023)
  • Research Focus: AI Assurance, ML Security, Vulnerability Assessment

🌐 Links

Personal Site LinkedIn Email


Bridging cybersecurity research with practical AI security solutions

Pinned Loading

  1. CountingMetrics CountingMetrics Public

    This probe connects to a remote AWS server via SSH to download machine learning models and analyze their parameters. It supports different model types, counting parameters for ONNX models and tree …

    Python 1

  2. IcaDevops IcaDevops Public

    This script applies Independent Component Analysis (ICA) to a dataset to extract independent components and evaluate their quality using kurtosis. The kurtosis values help determine whether the da…

    Python 1

  3. JA3Block-Check JA3Block-Check Public

    Python program that calculate ja3, match with your JA3Blacklist and can block malicious connections. Type -help for INFO.

    Python 1

  4. MembershipInferenceCheck MembershipInferenceCheck Public

    This probe evaluates ONNX models to determine their vulnerability to membership inference attacks, a type of privacy threat. It analyzes the model's behavior probabilistically to assess the risk of…

    Python 1

  5. MostImportantVarEvaluation MostImportantVarEvaluation Public

    This probe analyzes datasets to identify the most important and potentially sensitive variables using mutual information and feature importance techniques. It helps ensure fairness in machine learn…

    Python 1