Security & Privacy of AI: Challenges & Solutions for Machine Learning Applications

Datum en locatie volgen later
Vlaamse AI Academie & KU Leuven

Programma

Part I Introduction (2-3h)

  • Course objectives, Outlook and Approach
  • Introduction to ICT security: basics and essentials, broad overview
  • Requirements for AI-centric applications, application architectures, known challenges

Part II: Security and Privacy Engineering (2-3h)

  • Threat Modeling and Risk Assessment
  • Architectures and blueprints
  • Regulations and compliance

Part III: Building Blocks: Techniques for data protection (2-3h)

  • Data storage in the new century
  • Data processing – an overview of cryptographic protection
  • MPC intro and applications
  • FHE state-of-the-art and implications
  • Protection of data in motion

Part IV: Security and Privacy posture of an ML architecture (2-3h)

  • Attack surface – Data, Model, Process and data Flow
  • Threat modelling for AI and known challenges
  • Adversaries and their capabilities
  • Architectural and algorithmic defenses
  • Security and privacy challenges in Federated Machine Learning

Part V: Adversarial Machine Learning(2h)

  • Vulnerabilities in ML systems – overview of ML specific attacks
  • Protection against adversarial examples
  • Evaluation and Robustness of solutions
  • Privacy aspects

Part VI: Experience Reports (2h) (Selected Topics)

  • Face recognition revisited
  • Biometrics and challenges
  • Anomaly detection

Part VII Workshop (2h)

  • AI applications that are sufficiently articulated to be discussed in a workshop

Organisatie

Deze cursus wordt georganiseerd door:

  • Dr. Davy Preuveneers, DistriNet (Gedistribueerde & Veilige Software) Departement Computerwetenschappen, KU Leuven
  • Prof. Dr. Wouter Joosen, DistriNet, Departement Computerwetenschappen, KU Leuven
  • Prof. Bart Preneel, COSIC (Computer Security & Industrial Cryptography, ESAT, KU Leuven.
i.s.m.

Praktisch

  • datum nog te bepalen
  • Locatie nog te bepalen
  • Contact: laura.padula@vlaamse-ai-academie.be
  • Taal: Engels
  • Doelgroep: onderzoekers met kennis van machine learning

Inschrijven?

  • Prijs: nog te bepalen

Klaar om te starten?

Deze opleiding is in opbouw, zodra we meer informatie hebben, vullen we de pagina aan.