Search result for:
AISTATS | 2022
Learning from Multiple Noisy Partial Labelers
Abstract
This work enables users to create partial labelers that output subsets of possible class labels would greatly expand the expressivity of programmatic weak supervision.