School on Applied AI for Sustainable Development

Administration

Description

An ICTP meeting

The School covers key aspects of Applied AI, focusing on designing, developing, and deploying practical solutions to real-world problems. It aims to equip scientists, primarily from developing countries, to tackle local challenges and contribute to global sustainability goals.

Embedded devices running ML models are becoming increasingly powerful, evolving from Cloud-based AI to EdgeAI. This shift enables the creation of more complex, sustainable, and real-time AI solutions for applications such as environmental monitoring, particle detection and classification, radiation monitoring, smart agriculture, waste management, and beyond.

The School includes hands-on exercises, group discussions, and case studies to reinforce key concepts. Participants will be introduced to open-source methods, software tools, and hardware platforms relevant to real-world deployment.

The School is linked to multiple MSCA EU-funded projects (ROBUST / AIAS / ANTIDOTE), and the Annual Workshop of the the the Rural Environmental Monitoring via ultra wide-ARea networKs And distriButed federated LEarning (REMARKABLE) EU project.

Topics

Foundations of AI & Sustainable Development Goals (SDGs)
Sector-Specific AI Tools and Technologies
Introduction to TinyML and EdgeAI
Data cleaning
Heterogeneous Computing
Fundamentals of Applied ML
Sensors and Data Collection
ML and Model Compression for FPGA-based EdgeAI Devices
Edge AI for the Internet of Things: Insights from Real-World Projects
Ethical & Sustainable AI

Grants: A limited number of grants are available to support the attendance of selected participants, with priority given to participants from developing countries. There is no registration fee.

Material

Group photo Poster Preliminary list of Participants

School on Applied AI for Sustainable Development | (smr 4210)

Go to day

Monday, 23 March 2026

08:45 - 09:00

08:45 Registration formalities 15'

REGISTRATION: Upon arrival, Visitors not staying in the ICTP Guest Houses, are kindly requested to complete registration formalities at the Adriatico Guest House (Lower level). The Registration Desk will be open in the morning starting from 8:45.

09:00 - 14:00

09:00 Opening Remarks & School Outline 30'

Speaker:	Maria Liz CRESPO (ICTP, Italy), Marco ZENNARO (ICTP, Italy)
Material:	Slides

09:30 Edge AI for the Internet of Things: Insights from Real-World Projects 45'

Speaker:	Dejan VUKOBRATOVIC (FTN-UNS, Serbia)
Material:	Slides

10:15 Coffee Break 30'

10:45 AI-Enabled Semantic and Goal-Oriented Communications for the Internet of Things 1h0'

Speaker:	Dejan VUKOBRATOVIC (FTN-UNS, Serbia)
Material:	Slides

11:45 Introduction to Machine Learning and Edge AI 45'

Speaker:	Romina MOLINA (ICTP, Italy)
Material:	Slides

12:30 Lunch Break 1h30'

14:00 - 17:30 AI Methods for Predictive Maintenance in Power Electronics
```
Speakers: Jacopo  FERRETTI (Bologna University, Italy), Mariano NERONE (Bologna University, Italy), Igor VALIC (Bologna University, Italy)
```
- 14:00 Understanding System Behavior and Failure Mechanisms in Power Electronics 45'
  
  Material: Slides
- 14:45 Hardware Setup and Data Acquisition for Power Electronics Monitoring 45'
- 15:30 Coffee Break 15'
- 15:45 Estimating Remaining Useful Life (RUL) Using AI Techniques 1h45' ( InfoLab )

Tuesday, 24 March 2026

09:00 - 14:00

09:00 When Learning Meets the Channel: Edge AI through Split and Semantic Design 1h0'

Speaker:	Vukan NINKOVIC (Technical Sciences University of Novi Sad, Serbia)
Material:	Slides

10:00 Group Photo 5'
10:05 Coffee Break 25'

10:30 AI for Sustainable Development 1h0'

Speaker:	Joao Paulo PITA DA COSTA (IRCAI, Slovenia)
Material:	Slides

11:30 Responsible AIoT and TinyML 1h0'

Speaker:	Joao Paulo PITA DA COSTA (IRCAI, Slovenia)
Material:	Slides

12:30 Lunch Break 1h30'

14:00 - 17:15 Smart Trapping System for Dacus Flies

Speakers: Ioannis KARYDIS (Ionian University, Greece), ERASMUS Christopher Stefan (PDMFC, Spain)

14:00 Collecting Data from Sensors for Pest Monitoring 45'

Material:

Slides

14:45 Hardware Components of the Smart Trapping System 45'

Material: Slides
15:30 Coffee Break 15'
15:45 Optical Identification of Dacus Flies 1h30' ( InfoLab )

Wednesday, 25 March 2026

09:00 - 17:45 Machine Learning and Model Compression on FPGA-Based Heterogeneous Devices

Speakers: Romina MOLINA (ICTP, Italy), Maynor Giovanni BALLINA ESCOBAR (ICTP, Italy), Maria Liz CRESPO (ICTP, Italy)

09:00 Fundamentals of Applied Machine Learning 45'

Material: Slides
09:45 Model Optimization and Compression Techniques 45'

Material: Slides
10:30 Coffee Break 30'

11:00 Lab 1: Training Machine Learning Models 1h30'

Material:

Link

Slides

11:00 Lab 2: Applying Model Compression Techniques 1h30'

Material: Slides
12:30 Lunch Break 1h30'
14:00 Using FPGAs to Accelerate Machine Learning Algorithms 45'

Material: Slides
14:45 HyperFPGA: Heterogeneous Computing on MPSoC-FPGA 45'

Material: Slides
15:30 Coffee Break 15'
15:45 Lab 3: Deploying Machine Learning Models on Heterogeneous MPSoC-FPGA Devices 2h0'

Material: Slides

Thursday, 26 March 2026

09:00 - 14:00 Introduction to EU-Funded AI Research Projects

09:00 REMARKABLE Project Overview 10'

Speaker:	Paolo CALCIATI (PDM Espana Proyectos y Soluciones informaticas s.l.)
Material:	Slides

09:10 REMARKABLE Stellenbosch: Dendrometer-Based Tree Growth Analytics 10'

Speaker: Christopher ERASMUS (PDMFC Spain)

09:20 R-PODID: AI Model for Motor Health Prediction and Network Anomaly Detection 10'

Speaker:	Teresa BRILHA DE SOUSA (PDMFC, Portugal)
Material:	Slides

09:30 Agent-Based Autonomous UAV Navigation in GPS-Denied Scenarios 10'

Speaker:	Carlos Alexander OSORIO QUERO (National Institute of Astrophysics Optics and Electronics (INAOE), Mexico)
Material:	Slides

09:40 Machine Learning–Based Multi-Data Fusion on FPGA-SoC for Smart Agriculture 10'

Speaker: Maynor Giovanni BALLINA ESCOBAR (ICTP, Italy)
09:50 Question and Answers 10'
10:00 Coffee Break 30'
10:30 AI + IoT: new solutions emerging 2h0'

Speaker: Srdjan KRČO (DunavNET doo, Serbia)
12:30 Lunch Break 1h30'

14:00 - 17:15 Edge AI and TinyML: Concepts and Applications
```
Speakers: Pietro MANZONI (Universitat Politècnica de València, Spain), Marco ZENNARO (ICTP, Italy)
```
- 14:00 Introduction to TinyML and Data Ingestion in IoT 45'
  
  Material: Link
- 14:45 Working with Xiao Sense: Hardware Basics and Initial Applications 45'
  
  Material: Link
- 15:30 Coffee Break 15'
- 15:45 Designing an Image Classification Application with the Xiao Sense 1h30' ( InfoLab )
  
  Material: Link

Friday, 27 March 2026

09:00 - 14:00 Deep Learning for Computer Vision

09:00 Deep Learning Techniques for Computer Vision 1h0'

Speaker: João Pedro CARVALHO (Universidade de Lisboa, Portugal)
10:00 Coffee Break 30'
10:30 Deep learning use case 1h30' ( InfoLab )

Speaker: João Pedro CARVALHO (Universidade de Lisboa, Portugal)

12:00 Distribution of Certificates and Closing Remarks 30'

Speaker:

Maria Liz CRESPO (ICTP, Italy), Marco ZENNARO (ICTP, Italy), Paolo CALCIATI (PDM Espana Proyectos y Soluciones informaticas s.l.) noname

12:30 Lunch Break 1h30'

14:00 - 17:15
- 14:00 Visit to ICTP STI Labs 2h30'
  
  Material: Labs Map
- 16:45 Coffee Break 30' ( Adriatico Cafeteria )