Number of the records: 1
DEnFi: Deep Ensemble Filter for Active Learning
- 1.
SYSNO ASEP 0531426 Document Type V - Research Report R&D Document Type The record was not marked in the RIV Title DEnFi: Deep Ensemble Filter for Active Learning Author(s) Ulrych, Lukáš (UTIA-B)
Šmídl, Václav (UTIA-B) RID, ORCIDIssue data Prague: ÚTIA AV ČR, v.v.i, 2020 Series Research Report Series number 2383 Number of pages 10 s. Publication form Print - P Language eng - English Country CZ - Czech Republic Keywords Deep Ensembles ; uncertainty ; neural networks Subject RIV IN - Informatics, Computer Science OECD category Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) R&D Projects GA18-21409S GA ČR - Czech Science Foundation (CSF) Institutional support UTIA-B - RVO:67985556 Annotation Deep Ensembles proved to be a one of the most accurate representation of uncertainty for deep neural networks. Their accuracy is beneficial in the task of active learning where unknown samples are selected for labeling based on the uncertainty of their prediction. Underestimation of the predictive uncertainty leads to poor exploration of the method. The main issue of deep ensembles is their computational cost since multiple complex networks have to be computed in parallel. In this paper, we propose to address this issue by taking advantage of the recursive nature of active learning. Specifically, we propose several methods how to generate initial values of an ensemble based of the previous ensemble. We provide comparison of the proposed strategies with existing methods on benchmark problems from Bayesian optimization and active classification. Practical benefits of the approach is demonstrated on example of learning ID of an IoT device from structured data using deep-set based networks. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2021
Number of the records: 1