2018/01/01 by Fayeem Aziz, Aaron S. W. Wong, James S. Welsh +1
Computer Science · Mathematics · #Acoustics #Algorithm #Artificial intelligence #Computer science #Dimensionality reduction #Double pendulum #Dynamics (music) #Evolutionary Algorithms and Applications #Feature (linguistics) #Image (mathematics) #Image Processing and 3D Reconstruction #Inverted pendulum #Manifold (fluid mechanics) #Manifold alignment #Music and Audio Processing #Noise (video) #Nonlinear dimensionality reduction #Pattern recognition (psychology) #Pendulum #Physics #Space (punctuation) #cs.LG #msc:57-06 #stat.ML
paper · pdf · doi:10.1007/978-3-030-04239-4_7
The final version will appear in ICONIP 2018. A DOI identifier to the final version will be added to the preprint, as soon as it is available
openalex publication_date 2018/01/01 · arxiv created 2018/09/20 · arxiv updated 2018/11/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
This study presents the results of a series of simulation experiments that evaluate and compare four different manifold alignment methods under the influence of noise. The data was created by simulating the dynamics of two slightly different double pendulums in three-dimensional space. The method of semi-supervised feature-level manifold alignment using global distance resulted in the most convincing visualisations. However, the semi-supervised feature-level local alignment methods resulted in smaller alignment errors. These local alignment methods were also more robust to noise and faster than the other methods.