Description |
An ICTP meeting in person This workshop aims to develop sustainable university programs in embedded machine learning (also known as TinyML). We will do so by first bridging the gap between leveraging standard open-access course materials and adapting these materials for local and regional contexts. We will also share best practices and learnings that our network has gained through their implementation of these adaptations over the past few years. We will then augment these courses by exploring the development of university research programs centered on embedded machine learning. We will discuss how to leverage these programs to provide students and faculty with extra-curricular experience and develop new technologies and publications in the field of embedded machine learning. We will also explore how these research agendas can be used to support scientific applications and sustainable development. 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. |
Workshop on Widening Access to TinyML Network by Establishing Best Practices in Education | (smr 3851)
Go to day
-
-
08:30 - 09:15
-
08:30
Registration formalities
45'
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 from 8.30 to 09.15.
-
08:30
Registration formalities
45'
-
09:15 - 10:30
-
09:15
Workshop Opening and Call to Action
15'
Speaker: Marco ZENNARO (ICTP, Italy) Material: Slides -
09:30
Introductions and Sharing of Attendees
30'
Speaker: Marco ZENNARO (ICTP, Italy) - 10:00 Coffee Break 30'
-
09:15
Workshop Opening and Call to Action
15'
-
10:30 - 17:00
Experience Sessions - Long TinyML course
-
10:30
Hardware Architectures for Embedded and Edge AI
30'
Speaker: Manuel ROVERI (Politecnico of Milan) Material: Slides -
11:00
An Undergrad Engineering Course on Project Development
30'
Speaker: Marcelo ROVAI (UNIFEI IESTI, Brazil) Material: Slides -
11:30
Launching TinyML edX and Long Term Support
30'
Speaker: Brian PLANCHER (Barnard College, Columbia University) Material: Slides - 12:00 Lunch 1h30'
-
13:30
Experiences in Teaching TinyML to (Under)Graduate Students
30'
Speaker: Jesus LOPEZ (Universidad Autonoma de Occidente, Colombia) Material: Slides -
14:00
Deprecation, Client Side, and TinyMLjs
30'
Speaker: Jeremy ELLIS (Mission Secondary School, Canada) Material: Slides -
14:30
Experience Session Reflections and Lessons Learned
45'
- 15:15 Coffee Break 25'
-
15:40
Experience Session Reflections and Lessons Learned (Continued)
1h20'
-
10:30
Hardware Architectures for Embedded and Edge AI
30'
-
08:30 - 09:15
-
-
09:00 - 13:45
TinyML Community Building
-
09:00
An Overview of Edge Impulse and its Latest Features
55'
Speaker: Alessandro GRANDE (Edge Impulse) Material: Slides - 09:55 Coffee break 15'
-
10:10
Academia-Industry Partnerships from the Perspective of the tinyML Foundation and a Call to Action for tinyMLedu
55'
Speaker: Evgeni GOUSEV (TinyML Foundation) Material: Slides -
11:05
Is open source all that is needed to create a good education programme? Three experiences in designing courses for massive adoption
55'
Speaker: David CUARTIELLES (Malmo University) Material: Slides - 12:00 Lunch break 1h15'
-
13:15
Designing Open Training Materials
30'
Speaker: Marcus RUB (Hahn-Schickard-Gesellschaft für angewandte Forschung e.V.) and Thomas AMBERG (University of Applied Sciences and Arts Northwestern) Material: Slides
-
09:00
An Overview of Edge Impulse and its Latest Features
55'
-
13:45 - 17:00
Experience Sessions - Short TinyML courses
-
13:45
TinyML Course at ITU
24'
Speaker: Sebastian BUETTRICH (University of Oregon, NSRC and ICTP, Italy) Material: Slides -
14:09
TinyML Teaching Experience
24'
Speaker: Solomon Gizaw TULU (Addis Ababa University) Material: Slides -
14:33
TinyML teaching experience and supervising capstone projects
24'
Speaker: Ronald CRIOLLO (Escuela Superior Politécnica del Litoral, Ecuador) Material: Slides -
14:57
The challenging first steps of graduate students on TinyML
24'
Speaker: Diego MENDEZ CHAVES (ICTP, Italy) Material: Slides -
15:21
Micro-Credential Course on TinyML
24'
Speaker: Rosdiadee Bin NORDIN (UKM, Malaysia) Material: Slides - 15:45 Coffee Break 15'
-
16:00
Experience Session Reflections and Lessons Learned
1h0'
-
13:45
TinyML Course at ITU
24'
-
19:00 - 20:30
-
19:00
Welcome Reception
1h30'
(
Adriatico Guest House Terrace
)
All participants are cordially invited to the Welcome Reception
-
19:00
Welcome Reception
1h30'
(
Adriatico Guest House Terrace
)
-
09:00 - 13:45
TinyML Community Building
-
-
09:00 - 16:45
Technical Deep Dives + Hardware Demos
-
09:00
Keynote: Making Sense of the Wild
1h0'
Speaker: Eric PAN (Seeed Studio) Material: Slides - 10:00 Coffee break 15'
-
10:15
Technical Talk: From LoRa to the Cloud: Bridging Physical and Digital Worlds
1h0'
Speaker: Pietro MANZONI (Universidad Politecnica de Valencia) -
11:15
Research Talk: Benefits and Challenges of using Low Cost Weather Stations
1h0'
Speaker: Paul KUCERA (UCAR/COMET) - 12:15 Lunch break 1h30'
-
13:45
Research Talk: Monitoring mosquitoes of public health importance with TinyML
1h0'
Speaker: Cyril CAMINADE (ICTP) - 14:45 Coffee break 30'
-
15:15
Hardware Demo
1h30'
Speaker: Marcelo ROVAI (UNIFEI IESTI) and Jose Antonio BAGUR (Arduino)
-
09:00
Keynote: Making Sense of the Wild
1h0'
-
09:00 - 16:45
Technical Deep Dives + Hardware Demos
-
-
09:00 - 14:00
Research in Embedded ML
-
09:00
Research Talks Part I
1h0'
Neena GOVEAS (TinyML research and human health monitoring) Jose Antonio BAGUR (Anomaly Detection Course)
- 10:00 Coffee break 15'
-
10:15
Research Talks Part II
2h0'
Laila KAZIMIERSKI (Animal tracking) Milan LUKIC (Lightweight digit recognition in utility metering, Anomaly detection in logistics asset tracking, Detection of fungal disease outbreak risk in agriculture) Jackline TUM (leveraging TinyML for illegal Logging detection) Halleluyah AWORINDE (Leveraging TinyML for vocalization signal-based Poultry Health Management) Brian PLANCHER (ML Sensors and Environmental Impact of TinyML)
- 12:15 Lunch break 1h30'
-
09:00
Research Talks Part I
1h0'
-
13:45 - 16:45
White Paper on Embedded ML University Program Design
Brian PLANCHER facilitator
-
13:45
Group work
45'
in Groups outline the key points for the following sections: Courses that can (and need) to be taught, best practices on sharing educational materials, steps toward building a research program, curriculum diagrams building off of (or centered around TinyML), etc.
-
14:30
Group work
15'
shuffle the groups up and have them keep working for cross pollination (we can random the groups ahead of time)
- 14:45 Coffee break 30'
-
15:15
Group work
30'
last random shuffle and group work
-
15:45
Group work
45'
report back on final landing points for the various sections and discuss as a large group. Hopefully we can all align on the outline of the paper for each section.
-
13:45
Group work
45'
-
17:00 - 20:00
- 17:00 SciFabLab visit 3h0'
-
09:00 - 14:00
Research in Embedded ML
-
-
09:00 - 16:30
Growing the 4D Network and Increasing Collaboration
-
09:00
Keynote: Teaching TinyML in ARM Laboratories
1h0'
Speaker: Stephen OZOIGBO (ARM) - 10:00 Coffee break 30'
-
10:30
Future of Embedded ML
2h0'
(Group Work, Brian PLANCHER facilitator) Where can we go from here? On-device learning? EE? CE? CS? AI? Connections to other domains? Robotics?
- 12:30 Lunch break 1h30'
-
14:00
Feedback
1h0'
how does this all fit in your curriculum? What can be done to improve your local experience and effectiveness?
-
15:00
Collaboration Discussion
1h0'
how can we better collaborate on teaching and research – what systems do we need to put in place? Can we leverage the learnings in LATAM to improve this elsewhere?
-
16:00
Closing ceremony
30'
-
09:00
Keynote: Teaching TinyML in ARM Laboratories
1h0'
-
09:00 - 16:30
Growing the 4D Network and Increasing Collaboration