# Manifold Alignment Determination: finding correspondences across different data views

Andreas Damianou, Amazon Research Cambridge
Neil D. Lawrence, Amazon Research Cambridge and University of Sheffield
Carl Henrik Ek, University of Bristol

#### Abstract

We present Manifold Alignment Determination (MAD), an algorithm for learning alignments between data points from multiple views or modalities. The approach is capable of learning correspondences between views as well as correspondences between individual data-points. The proposed method requires only a few aligned examples from which it is capable to recover a global alignment through a probabilistic model. The strong, yet flexible regularization provided by the generative model is sufficient to align the views. We provide experiments on both synthetic and real data to highlight the benefit of the proposed approach.

  @TechReport{manifold-alignment-determination, title = {Manifold Alignment Determination: finding correspondences across different data views}, author = {Andreas Damianou and Neil D. Lawrence and Carl Henrik Ek}, year = {2017}, institution = {}, month = {00}, edit = {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2017-01-12-manifold-alignment-determination.md}, url = {http://inverseprobability.com/publications/manifold-alignment-determination.html}, abstract = {We present Manifold Alignment Determination (MAD), an algorithm for learning alignments between data points from multiple views or modalities. The approach is capable of learning correspondences between views as well as correspondences between individual data-points. The proposed method requires only a few aligned examples from which it is capable to recover a global alignment through a probabilistic model. The strong, yet flexible regularization provided by the generative model is sufficient to align the views. We provide experiments on both synthetic and real data to highlight the benefit of the proposed approach.}, key = {Damianou:mad16}, linkpdf = {https://arxiv.org/abs/1701.03449.pdf}, OPTgroup = {} }
 %T Manifold Alignment Determination: finding correspondences across different data views %A Andreas Damianou and Neil D. Lawrence and Carl Henrik Ek %B %D %F manifold-alignment-determination %P -- %R %U http://inverseprobability.com/publications/manifold-alignment-determination.html %X We present Manifold Alignment Determination (MAD), an algorithm for learning alignments between data points from multiple views or modalities. The approach is capable of learning correspondences between views as well as correspondences between individual data-points. The proposed method requires only a few aligned examples from which it is capable to recover a global alignment through a probabilistic model. The strong, yet flexible regularization provided by the generative model is sufficient to align the views. We provide experiments on both synthetic and real data to highlight the benefit of the proposed approach. 
 TY - CPAPER TI - Manifold Alignment Determination: finding correspondences across different data views AU - Andreas Damianou AU - Neil D. Lawrence AU - Carl Henrik Ek PY - 2017/01/12 DA - 2017/01/12 ID - manifold-alignment-determination SP - EP - L1 - https://arxiv.org/abs/1701.03449.pdf UR - http://inverseprobability.com/publications/manifold-alignment-determination.html AB - We present Manifold Alignment Determination (MAD), an algorithm for learning alignments between data points from multiple views or modalities. The approach is capable of learning correspondences between views as well as correspondences between individual data-points. The proposed method requires only a few aligned examples from which it is capable to recover a global alignment through a probabilistic model. The strong, yet flexible regularization provided by the generative model is sufficient to align the views. We provide experiments on both synthetic and real data to highlight the benefit of the proposed approach. ER - 
 Damianou, A., Lawrence, N.D. & Ek, C.H.. (2017). Manifold Alignment Determination: finding correspondences across different data views.:-