Year: 2020
Type: preprint
Abstract: We study a pool-based sequential active learning (AL), in which one sample is queried at each time from a large pool of unlabeled data according to a selection criterion. For this framework, we propose two selection criteria, named expected-kernel-discrepancy (EKD) and expected-kernel-loss (EKL), by leveraging the particular structure of multiple kernel learning (MKL). Also, it is identified that the proposed EKD and EKL successfully generalize the concepts of popular query-by-committee (QBC) and expected-model-change (EMC), respectively. Via experimental results with real-data sets, we verify the effectiveness of the proposed criteria compared with the existing methods.
Source: arXiv (Cornell University)
Authors: Chae, Jeongmin, Hong, Songnam
Institutions: University of Southern California, Hanyang University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3094046485