SOINN+, a Self-Organizing Incremental Neural Network for Unsupervised Learning from Noisy Data Streams

作者：

Highlights：

• We present a new self-organizing incremental neural network: SOINN+.

• SOINN+ is for unsupervised, continuous learning.

• SOINN+ can detect clusters in noisy, evolving data streams.

• SOINN+ mitigates catastrophic forgetting.

摘要

•We present a new self-organizing incremental neural network: SOINN+.•SOINN+ is for unsupervised, continuous learning.•SOINN+ can detect clusters in noisy, evolving data streams.•SOINN+ mitigates catastrophic forgetting.

论文关键词：Continuous learning,Catastrophic forgetting,Self-organizing incremental neural network,Non-stationary data,Data stream,Concept drift

论文评审过程：Received 15 July 2019, Revised 11 October 2019, Accepted 31 October 2019, Available online 2 November 2019, Version of Record 8 November 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.113069