The Trustworthy ML Initiative celebrates its one year anniversary with this special event, held jointly with Montreal AI Ethics Institute.
To achieve the promise of AI for societal impact, black-box models must not only be "accurate" but also satisfy trustworthiness properties that facilitate open collaboration and ensure ethical outcomes. The purpose of this un-symposium is to discuss the interdisciplinary topics of robustness, fairness, privacy, and ethics of AI tools. In particular, we want to highlight the significant gap in deploying these AI models in practice when the stakes are high for commercial applications of AI where millions of human lives are at risk. While these challenges look critical, we believe we can overcome this with the collective effort of researchers, stakeholders, and domain experts.
Agenda (Eastern Time):
10:30am -- Opening remarks: What is the invisible elephant in the room?
10:45am -- Panel on "Interdisciplinary Research in Trustworthy ML -- Challenges and Way Forward". Panelists: Danielle Belgrave (DeepMind), Tom Dietterich (Oregon State Univ), Kush Varshney (IBM Research). Moderator: Subho Majumdar (Splunk)
11:45am -- Break
12:00 -- Townhall on "Practical Challenges of Applying Trustworthy ML in Industry". Panelists: Stella Biderman (EleutherAI), Cristian Canton Ferrer (Facebook), Krishnaram Kenthapadi (Amazon). Moderator: Abhishek Gupta (Montreal AI Ethics Institute)
1:00 -- Break. Adjourn to separate Zoom link https://us02web.zoom.us/j/84698021270?pwd=cjBXaWVuRlRuZUI1VGczcXJTZVZmQT09
1:15 to 3pm -- Social with breakout rooms. Host: Chirag Agarwal (Harvard Univ)
#1 -- Interpretability. Host: Chhavi Yadav (UCSD), Masa Sweidan (McGill University)
#2 -- Is unfairness a security risk? Host: Mohammad Yaghini (U Toronto)
#3 -- Out-of-distribution robustness. Host: Haohan Wang (CMU)
#4 -- Fairness. Host: Marta Lemanczyk (Hasso Plattner Institute), Marianna Ganapini (Union College)
#5 -- Causal inference. Host: Maggie Makar (U Michigan)