Parametric Herding

A parametric version of herding is formulated. The nonlinear mapping between consecutive time slices is learned by a form of self-supervised training. The resulting dynamical system generates pseudo-samples that resemble the original data. We show how this parametric herding can be successfully used to compress a dataset consisting of binary digits. It is also verified that high compression rates translate into good prediction performance on unseen test data.

• Type: Journal
• Year: 2010
• Where: JMLR
• Authors: Yutian Chen, Max Welling
• Keywords: Consecutive Time Slices, JMLR 2010, Parametric Version, Unseen Test Data