4hr/day schedule (2hr30 on Thursday 14)
Start of the workshop:
5pm - 9pm Israel time 4pm - 8pm CET 3pm - 7pm GMT
9am - 1pm CST 10am - 2pm AST 7am - 11am PT
Time slots below show duration SINCE START (00:00).
Presentations will be done live for this session.
00:00 - 00:15 Welcomes by the Lorentz Center and organising committee
00:15 - 02:15 Presentations of industrial needs and legal context for AI robustness
Vahid Hashemi (Audi) for automotive
Romaric Redon (Airbus) for aeronautics
Guillaume Soudain (EASA) for aeronautical certification
Mireille Hildebrandt (VBU) for legal context
02:15 - 02:30 Break
--
Panel session
Themes: Aligned/diverging industrial AI robustness challenges. AI shortcomings. Short, mid-, long-term solutions. Evolution of laws and their implications for the industry.
02:30 - 03:15 Panel discussion with presenter
03:15 - 04:00 Interactions between panel and participants
To follow this session, participants are invited to look at the pre-recorded talks on ‘Empirical robustness’ here.
00:00 - 01:00 Talk recap (1min/talk) & Q&A on pre-recorded reviews/talks
--
Break-out session: Panel/sub-groups detailed here
Themes: Adversarial attacks: a true threat to real-life applications? - How can empirical verification be used in an industrial qualification/certification framework? - Where lies the balance between AI architecture complexity and vulnerability to attacks?
01:00 - 01:30 Introduction (panel)
01:30 - 02:00 Break-out session / Brainstorming - Participants separated in discussion groups (x 3-4)
02:00 - 02:15 Break
02:15 - 03:15 Debriefing break-out session & panel discussion
--
03:15 - 04:00 Live: Library/SW demonstration: Adversarial ML (7:15pm CET)
Beat Buesser & Mathieu Sinn: IBM Adversarial Robustness ToolBox
To follow this session, participants are invited to look at the pre-recorded talks on ‘Formal robustness’ here.
00:00 - 01:00 Talk recap (1min/talk) & Q&A on pre-recorded reviews/talks
--
Break-out session: Panel/sub-groups detailed here
Themes: Scalability - Complementarity of techniques
01:00 - 01:30 Introduction (panel)
01:30 - 02:00 Break-out session / Brainstorming - Participants separated in discussion groups (x 3-4)
02:00 - 02:15 Break
02:15 - 03:15 Debriefing break-out session & panel discussion
--
03:15 - 04:00 Live: Library/SW demonstrations: Formal methods (7:15pm CET) - Run in parallel
Aleksandar Zeljic: Marabou
Huan Zhang: auto_LIRPA
00:00 - 01:00 Panel discussion 1 - Theme: 'How will current and future research on robustness assessment techniques solve AI industrial challenges?'
01:00 - 02:00 Panel discussion 2 - Theme: 'Ways forward & research road map' + 'Future collaboration: Favoured means & platforms?'
02:00 - 02:30 Panel discussion 3 - Theme: 'The workshop legacy'