Improving the performance of bagging ensembles for data streams through mini-batching - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Information Sciences Année : 2021

Improving the performance of bagging ensembles for data streams through mini-batching

Résumé

Often, machine learning applications have to cope with dynamic environments where data are collected in the form of continuous data streams with potentially infinite length and transient behavior. Compared to traditional (batch) data mining, stream processing algorithms have additional requirements regarding computational resources and adaptability to data evolution. They must process instances incrementally because the data’s continuous flow prohibits storing data for multiple passes. Ensemble learning achieved remarkable predictive performance in this scenario. Implemented as a set of (several) individual classifiers, ensembles are naturally amendable for task parallelism. However, the incremental learning and dynamic data structures used to capture the concept drift increase the cache misses and hinder the benefit of parallelism. This paper proposes a mini-batching strategy that can improve memory access locality and performance of several ensemble algorithms for stream mining in multi-core environments. With the aid of a formal framework, we demonstrate that mini-batching can significantly decrease the reuse distance (and the number of cache misses). Experiments on six different state-of-the-art ensemble algorithms applying four benchmark datasets with varied characteristics show speedups of up to 5X on 8-core processors. These benefits come at the expense of a small reduction in predictive performance.

Dates et versions

hal-04468414 , version 1 (20-02-2024)

Identifiants

Citer

Guilherme Weigert Cassales, Heitor Murilo Gomes, Albert Bifet, Bernhard Pfahringer, Hermes Senger. Improving the performance of bagging ensembles for data streams through mini-batching. Information Sciences, 2021, 580, pp.260--282. ⟨10.1016/J.INS.2021.08.085⟩. ⟨hal-04468414⟩
5 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More