Abstract:In streaming data classification, most of the existing methods assume that all arrived evolving data are completely labeled. One challenge is that some applications where only small amount of labeled examples are available for training. Incremental semi-supervised learning algorithms have been proposed for regularizing neural networks by incorporating various side information, such as pairwise constraints or user-provided labels. However, it is hard to put them into practice, especially for non-stationary environments due to the effectiveness and parameter sensitivity of such algorithms. In this paper, we propose a novel incremental semi-supervised learning framework on streaming data. Each layer of model is comprised of a generative network, a discriminant structure and the bridge. The generative network uses dynamic feature learning based on autoencoders to learn generative features from streaming data which has been demonstrated its potential in learning latent feature representations. In addition, the discriminant structure regularizes the network construction via building pairwise similarity and dissimilarity constraints. It is also used for facilitating the parameter learning of the generative network. The network and structure are integrated into a joint learning framework and bridged by enforcing the correlation of their parameters, which balances the flexible incorporation of supervision information and numerical tractability for non-stationary environments as well as explores the intrinsic data structure. Moreover, an efficient algorithm is designed to solve the proposed optimization problem and we also give an ensemble method. Particularly, when multiple layers of model are stacked, the performance is significantly boosted. Finally, to validate the effectiveness of the proposed method, extensive experiments are conducted on synthetic and real-life datasets. The experimental results demonstrate that the performance of the proposed algorithms is superior to some state-of-the-art approaches.

Adaptive Learning for Weakly Labeled Streams

An Adaptive Learning Approach for Noisy Data Streams

Learning with Asynchronous Labels

A reliable adaptive prototype-based learning for evolving data streams with limited labels

Classifying noisy data streams

Semi-Supervised Streaming Learning with Emerging New Labels

Learning High-Dimensional Evolving Data Streams with Limited Labels

Online Active Learning for Drifting Data Streams

Learning Evolving Prototypes for Imbalanced Data Stream Classification with Limited Labels

Clustering-based Active Learning Classification towards Data Stream

Improved Adaptive Algorithm for Scalable Active Learning with Weak Labeler

Active Learning for Streaming Networked Data

Weakly Supervised Label Learning Flows

A Novel Data Stream Learning Approach to Tackle One-Sided Label Noise From Verification Latency

Streaming Label Learning for Modeling Labels on the Fly.

CSAL: Cost sensitive active learning for multi-source drifting stream

A Practical Approach to Classify Evolving Data Streams: Training with Limited Amount of Labeled Data

Adaptive Online Incremental Learning for Evolving Data Streams

Robust Ensemble Learning for Mining Noisy Data Streams.

Online Reliable Semi-supervised Learning on Evolving Data Streams

Incremental semi-supervised learning on streaming data.