📅 Date: May 5, 2026

🕘 Time: 09:00 – 14:00

📍 Location: Main Lecture Room, Jožef Stefan Institute (JSI), Ljubljana

This one-day training event brings together students, researchers, and young entrepreneurs to explore the scientific, technological, and policy aspects of trustworthy AI, data privacy, and security, combining a scientific tutorial with a policy discussion on emerging European regulation.

Agenda

09:00 – 09:10

Welcome and Opening Remarks

ERA Chair AutoLearn-SI and SLAIF Team

Scientific Tutorial Session

Knowledge Graph-Powered Decentralized Personalization

Presenters:

-- Asst. Prof. Oshani Seneviratne

Assistant Professor of Computer Science, Rensselaer Polytechnic Institute (RPI), Leader of the BRAINS Lab

-- Fernando Spadea

PhD Student, Rensselaer Polytechnic Institute (RPI)

09:10 – 09:25

Introduction

Tutorial objectives and structure

Overview of decentralized personalization

09:25 – 10:15

Personal Knowledge Graph Foundations

Topics include:

Limitations of centralized personalization systems

Personal Knowledge Graph (PKG) representations

Provenance, schema alignment, and user-centric modeling

Applications of PKGs in recommendation systems and domain knowledge graphs

10:15 – 10:30

Decentralized AI Architectures (Part I)

Federated learning for decentralized personalization

Privacy-preserving computation

10:30 – 10:45 ☕ Coffee Break

10:45 – 11:15

Decentralized AI Architectures (Part II)

Blockchain-based trust infrastructures

Applications in healthcare and intelligent decision-support systems

11:15 – 11:35

Semantic–Neural Personalization Approaches

Topics include:

Rule-guided knowledge graph adaptation

Behavioral alignment methods

Filter-bubble mitigation strategies

Applications in financial recommendation systems

11:35 – 12:00

Knowledge-Graph-Grounded RAG Pipelines

Topics include:

Parallel and multi-stage retrieval strategies

Building compact personalized contexts

Integration with large language models

Knowledge-graph-grounded retrieval-augmented generation

12:00 – 13:00 🍽 Lunch Break

13:00 – 13:30

Hands-On Exercise

Participants will use the provided Jupyter notebooks to:

Build a small Personal Knowledge Graph (PKG) using RDFLib

Run a toy federated personalization workflow using Flower

Execute knowledge graph–based retrieval for LLM prompting using LangChain

13:30 – 14:00

Policy Session: The EU AI Act

Presentation on key aspects and implications of the EU AI Act

Speaker:

-- Dr. Polona Pičman Štefančič

Ministry of Digital Transformation, Slovenia