Structured and Unstructured KBs

Danqi Chen, Rajarshi Das, Angela Fan, Sewon Min, Siva Reddy, Pat Verga

Please visit https://uskb-workshop.github.io/ for more details on the workshop schedule.

Schedule

0:00:00-
Kenneth Forbus, Northwestern University
0:30:16-
Philipp Cimiano, Bielefeld University
0:58:40-
Will Hamilton, McGill University / Mila
1:30:39-
Yunyao Li, IBM Research
2:01:21-
Fabio Petroni, Facebook AI Research
2:32:43-
Colin Raffel, University of North Carolina
3:08:54-
Panel discussion

Abstract:

Knowledge bases have been a key part of AI research and applications for decades. Traditionally, KBs were purely symbolic but more recently deep learning advances have raised the possibility of having purely unstructured/implicit neural KBs.

The objectives of this workshop is to bring together researchers exploring different paradigms of extracting, representing and applying knowledge. We will discuss knowledge spanning across Structured (tables, graphs, etc) to Unstructured (plain text), and Explicit (symbolic) to Implicit (parameterized).