Concept Drift Handling: A Domain Adaptation Perspective

EXPERT SYSTEMS WITH APPLICATIONS（2023）

Cited 7|Views10

Abstract

Data stream prediction is challenging when concepts drift, processing time, and memory constraints come into account. Concept drift refers to changes in data distribution over time that reduces prediction systems’ accuracy. We present a method for handling concept drift with a domain adaptation approach (CDDA) in a data stream. The proposed method passively deals with the concept drift by using the domain adaptation approaches with multiple sources while reducing the model execution time and memory consumption. We introduce two variants of CDDA to transfer the information in the multi-source windows to the target window: weighted multi-source CDDA and multi-source feature alignment CDDA. Then, we theoretically study the behavior of CDDA and find the generalization bound of CDDA for the data stream prediction problem. Moreover, an extensive set of experiments conducted on both synthetic and real-world data streams confirms the validity and excellent performance of the proposed approach. Our code is available at https://github.com/mahan66/cdda.

Translated text

Key words

Data stream prediction,Domain adaptation,Concept drift,Generalization bound,Uniform entropy number

求助PDF

上传PDF

Bibtex

AI Read Science

AI Summary

AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.

Example

Background

Key content

Introduction

Methods

Results

Related work

Fund

Key content

Pretraining has recently greatly promoted the development of natural language processing (NLP)
We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance

Upload PDF to Generate Summary

Must-Reading Tree

Example

Generate MRT to find the research sequence of this paper