Please use this identifier to cite or link to this item:
Title: Learning under concept drift with follow the regularized leader and adaptive decaying proximal
Authors: Huynh, Ngoc Anh
Ng, Wee Keong
Ariyapala, Kanishka
Keywords: Concept Drift
Decaying Rate
Issue Date: 2018
Source: Huynh, N. A., Ng, W. K., & Ariyapala, K. (2018). Learning under concept drift with follow the regularized leader and adaptive decaying proximal. Expert Systems with Applications, 96, 49-63.
Series/Report no.: Expert Systems with Applications
Abstract: Concept drift is the problem that the statistical properties of the data generating process change over time. Recently, the Time Decaying Adaptive Prediction (TDAP) algorithm1 was proposed to address the problem of concept drift. TDAP was designed to account for the effect of drifting concepts by discounting the contribution of previous learning examples using an exponentially decaying factor. The drawback of TDAP is that the rate of its decaying factor is required to be manually tuned. To address this drawback, we propose a new adaptive online algorithm, called Follow-the-Regularized-Leader with Adaptive Decaying Proximal (FTRL-ADP). There are two novelties in our approach. First, we derive a rule to automatically update the decaying rate, based on a rigorous theoretical analysis. Second, we use a concept drift detector to identify major drifts and reset the update rule accordingly. Comparative experiments with 14 datasets and 6 other online algorithms show that FTRL-ADP is most advantageous in noisy environments with real drifts.
ISSN: 0957-4174
DOI: 10.1016/j.eswa.2017.11.042
Rights: © 2017 Elsevier Ltd. This is the author created version of a work that has been peer reviewed and accepted for publication by Expert Systems with Applications, Elsevier Ltd. It incorporates referee’s comments but changes resulting from the publishing process, such as copyediting, structural formatting, may not be reflected in this document. The published version is available at: [].
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Journal Articles

Citations 20

Updated on Sep 6, 2020

Citations 20

Updated on Mar 4, 2021

Page view(s)

Updated on May 15, 2022

Download(s) 20

Updated on May 15, 2022

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.