Spherical Principal Curves
نویسندگان
چکیده
This paper presents a new approach for dimension reduction of data observed on spherical surfaces. Several techniques have been developed in recent years non-euclidean analysis. As pioneer work, (Hauberg 2016) attempted to implement principal curves Riemannian manifolds. However, this uses approximations process manifolds, resulting distorted results. study proposes project onto continuous curve construct Our lies the same line (Hastie and Stuetzle et al. 1989) that proposed euclidean space. We further investigate stationarity satisfy self-consistency The results real analysis simulation examples show promising empirical characteristics approach.
منابع مشابه
Principal Curves
Principal curves are smooth one-dimensional curves that pass through the middle of a p-dimensional data set, providing a nonlinear summary of the data. They are nonparametric, and their shape is suggested by the data. The algorithm for constructing principal curves starts with some prior summary, such as the usual principal-component li e. The curve in each successive iteration is a smooth or l...
متن کاملPrincipal Curves : Learning
and submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Computer Science) complies with the regulations of this University and meets the accepted standards with respect to originality and quality. The subjects of this thesis are unsupervised learning in general, and principal curves in particular. Principal curves were originally defined by Hastie [Has84...
متن کاملEstimations of Principal Curves
Principal curves are smooth curves that minimize the average squared orthogonal distance to each point in a data set. Fitting a principal curve is a maximum-likelihood technique for nonlinear regression in the presence of Gaussian noise on both x and y. We choose two definitions of principal curves in the literature and then present experimental results to discuss over them.
متن کاملSequential Principal Curves Analysis
This report includes all the technical details of the Sequential Principal Curves Analysis (SPCA) in a single document. SPCA is an unsupervised nonlinear and invertible feature extraction technique. The identified curvilinear features can be interpreted as a set of nonlinear sensors: the response of each sensor is the projection onto the corresponding feature. Moreover, it can be easily tuned f...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Pattern Analysis and Machine Intelligence
سال: 2021
ISSN: ['1939-3539', '2160-9292', '0162-8828']
DOI: https://doi.org/10.1109/tpami.2020.3025327