The ancient Greek word Agora refers to a public open space used for assemblies and markets. It captures the informal nature of our weekly meetings, a place where exchange of knowledge, ideas, and an engaging conversation takes place.
- Due to the COVID-19 situation, we will organize our meetings virtually for the foreseeable future
- If you wish to join our upcoming meetings, please subscribe to the AiTech Agora mailing list here, or just send an email with a subscription request to firstname.lastname@example.org. You will then start receiving weekly invitations containing the link to the virtual meeting room and the password
- If you haven't received the invitation through our mailing list and want to join a particular meeting at short notice, please email email@example.com
- Presentation slides from our past meetings are available here. We also record some of our online meetings, the past recordings are available here
May 19, 13:00 – 14:00 CEST
Jie Yang: ARCH: Know What Your Machine Doesn't Know
Despite their impressive performance, machine learning systems remain prohibitively unreliable in safety-, trust- and ethically sensitive domains. Recent discussions in different sub-fields of AI have reached the consensus of knowledge need in machines; few discussions, however, have touched upon the diagnosis of what knowledge is needed. In this proposal, I will present our ongoing work on ARCH, a human-in-the-loop, reasoning-based tool for diagnosing the unknowns of a machine learning system. ARCH leverages human intelligence to create domain knowledge required for a given task and to describe the internal behavior of a machine learning system; it infers the missing or incorrect knowledge of the system with the built-in probabilistic, abductive reasoning engine. ARCH is a generic tool that can be applied to machine learning in different contexts; in the talk, I will present several applications in which ARCH is currently being developed and tested, including health, finance, smart buildings, and conversational agents.
Jie Yang is Assistant Professor at the Web Information Systems (WIS) group of the Faculty of Engineering, Mathematics and Computer Science (EEMCS/EWI), Delft University of Technology. He co-leads the Delft Design@Scale AI Lab and at the WIS group, the research line on Crowd Computing and Human-Centered AI. Before joining TU Delft, Jie was Machine Learning Scientist at Amazon Research (Seattle), working on core AI algorithms for Alexa, and Senior Researcher at the eXascale Infolab, University of Fribourg - Switzerland. Jie's research lies at the intersection of Human Computation, Machine Learning, and Data Science. He is interested in developing methods and tools for building Human-Centered Machine Learning systems that are more interpretable and reliable by actively involving relevant stakeholders and domain experts throughout the lifecycle of the systems. He is currently serving as the Co-Editor for the Frontiers in AI journal "Human-Centered AI: Crowd Computing".
*If you wish to unsubscribe from this mailing list, please send an email with subject line “unsubscribe” to firstname.lastname@example.org.
Would like to join? Send us an email and we will forward the meeting details (aitech <> tudelft.nl).
- May 19: Jie Yang
May 19: 13:00-14:00
Jie Yang: ARCH: Know What Your Machine Doesn’t Know
- May 26: Pradeep Murukannaiah
May 26: 1300-1400 CEST
- June 9: Sebastian Kohler
June 9: 13:00-14:00
Sebastian Kohler: TBD