MiPPS: A generative model for multi-manifold clustering

Oluwasanmi Koyejo, Joydeep Ghosh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a generative model for high dimensional data consisting of intrinsically low dimensional clusters that are noisily sampled. The proposed model is a mixture of probabilistic principal surfaces (MiPPS) optimized using expectation maximization. We use a Bayesian prior on the model parameters to maximize the corresponding marginal likelihood. We also show empirically that this optimization can be biased towards a good local optimum by using our prior intuition to guide the initialization phase. The proposed unsupervised algorithm naturally handles cases where the data lies on multiple connected components of a single manifold and where the component manifolds intersect. In addition to clustering, we learn a functional model for the underlying structure of each component cluster as a parameterized hyper-surface in ambient noise. This model is used to learn a global embedding that we use for visualization of the entire dataset. We demonstrate the performance of MiPPS in separating and visualizing land cover types in a hyperspectral dataset.

Original languageEnglish (US)
Title of host publicationManifold Learning and Its Applications - Papers from the AAAI Fall Symposium, Technical Report
Pages18-25
Number of pages8
StatePublished - Dec 1 2009
Externally publishedYes
Event2009 AAAI FAll Symposium - Arlington, VA, United States
Duration: Nov 5 2009Nov 7 2009

Publication series

NameAAAI Fall Symposium - Technical Report
VolumeFS-09-04

Other

Other2009 AAAI FAll Symposium
CountryUnited States
CityArlington, VA
Period11/5/0911/7/09

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'MiPPS: A generative model for multi-manifold clustering'. Together they form a unique fingerprint.

  • Cite this

    Koyejo, O., & Ghosh, J. (2009). MiPPS: A generative model for multi-manifold clustering. In Manifold Learning and Its Applications - Papers from the AAAI Fall Symposium, Technical Report (pp. 18-25). (AAAI Fall Symposium - Technical Report; Vol. FS-09-04).