Deep structured energy based models for anomaly detection
Shuangfei Zhai, Yu Cheng, et al.
ICML 2016
Active learning is a promising way to efficiently build up training sets with minimal supervision. Most existing methods consider the learning problem in a pool-based setting. However, in a lot of real-world learning tasks, such as crowd-sourcing, the unlabeled samples, arrive sequentially in the form of continuous rapid streams. Thus, preparing a pool of unlabeled data for active learning is impractical. Moreover, performing exhaustive search in a data pool is expensive, and therefore unsuitable for supporting on-the-fly interactive learning in large scale data. In this paper, we present a systematic framework for stream-based multi-class active learning. Following the reinforcement learning framework, we propose a feedback-driven active learning approach by adaptively combining different criteria in a time-varying manner. Our method is able to balance exploration and exploitation during the learning process. Extensive evaluation on various benchmark and real-world datasets demonstrates the superiority of our framework over existing methods. Copyright 2013 ACM.
Shuangfei Zhai, Yu Cheng, et al.
ICML 2016
Yuan Luo, Yu Cheng, et al.
JAMIA
Zhengping Che, Yu Cheng, et al.
ICDM 2017
Zhaonan Sun, Soumya Ghosh, et al.
JAMIA Open