Timezone: »

HRN: A Holistic Approach to One Class Learning
Wenpeng Hu · Mengyu Wang · Qi Qin · Jinwen Ma · Bing Liu

Wed Dec 09 09:00 PM -- 11:00 PM (PST) @ Poster Session 4 #1220

Existing neural network based one-class learning methods mainly use various forms of auto-encoders or GAN style adversarial training to learn a latent representation of the given one class of data. This paper proposes an entirely different approach based on a novel regularization, called holistic regularization (or H-regularization), which enables the system to consider the data holistically, not to produce a model that biases towards some features. Combined with a proposed 2-norm instance-level data normalization, we obtain an effective one-class learning method, called HRN. To our knowledge, the proposed regularization and the normalization method have not been reported before. Experimental evaluation using both benchmark image classification and traditional anomaly detection datasets show that HRN markedly outperforms the state-of-the-art existing deep/non-deep learning models.

Author Information

Wenpeng Hu (Peking University)
Mengyu Wang (Peking University)
Qi Qin (Peking University)
Jinwen Ma (Peking University)
Bing Liu (Peking University)

More from the Same Authors