Spherical Embedding and Classification
Classifiers in Elliptic space In practical applications, we want to do some kind of learning on the data, for example classification • NN classifier is straightforward, as we can compute distances • In principle, we can implement any geometric classifier as we have a smooth metric space
A randomized sphere cover classifier | Proceedings of the …
This paper describes an instance based classifier, the randomised sphere covering classifier (αRSC), that reduces the training data set size without loss of accuracy when compared to nearest neighbour classifiers. The motivation for developing this algorithm is the desire to have a non-deterministic, fast, instance based classifier that ...
Spherical Classification of Remote Sensing Data
A task-specific hierarchically designed network with a spherical space classifier (HDSS-Net) to alleviate the issues of ship recognition with large size variation and a novel spherical spaceclassifier is proposed to expand the interclass margin and compress the intraclass feature distribution by fully taking advantage of the property of spherical space.
Margin setting algorithm for pattern classification via …
A generalized sphere classifier is presented by Lee during the adaption stage . Both the size and location of the spheres can be adjusted until they cover the correct class. The set covering machine (SCM) proposed by Marchand and Shawe-Taylor is a classifier that outputs a conjunction or disjunction of a set of spherical classifiers . Each of ...
A Randomized Sphere Cover Classifier | Semantic Scholar
An instance based classifier, the randomised sphere covering classifier (αRSC), that reduces the training data set size without loss of accuracy when compared to nearest neighbour classifiers is described. This paper describes an instance based classifier, the randomised sphere covering classifier (αRSC), that reduces the training data set size without …
david-dunson/SPAclassifier: Spherical Approximation Classifier
As a realization of LOcal Manifold Approximation (LOMA) classifier, SPherical Approximation (SPA) classifier is designed for classification when the feature in different classes have complex, nonlinear, overlapping and intersecting support, especially when the training data are limited.
Spherical Space Domain Adaptation With Robust Pseudo …
Adversarial domain adaptation (DA) has been an effective approach for learning domain-invariant features by adversarial training. In this paper, we propose a novel adversarial DA approach …
HDSS-Net: A Novel Hierarchically Designed Network With …
Finally, a novel spherical space classifier (SSC) is proposed to expand the interclass margin and compress the intraclass feature distribution by fully taking advantage of the property of …
Spherical Classification of Data, a New Rule-Based …
This paper presents a new rule-based classification method that partitions data under analysis into spherical patterns. The forte of the method is twofold. One, it exploits the efficiency of distance metric-based clustering to fast collect similar data into spherical patterns. The other, spherical patterns are each a trait shared among one type of data only, hence are …
A novel air-suction classifier for fresh sphere fruits in …
To improve the harvesting efficiency of the sphere fruits and reduce the workload of post-harvest classification, an air-suction sphere fruit classifier was designed in this paper, which can achieve the separation of high-quality fruit and common fruit through a pneumatic device according to the size and quality of the fruit when the cones are picked.
FULLY STEERABLE 3D SPHERICAL NEURONS
a spherical neuron (classifier), i.e., the hypersphere neuron (Banarer et al., 2003b) or its generalization for 3D input point sets — the geometric neuron (Melnyk et al., 2021) — can be turned into a steerable neuron. We prove that the aforementioned spherical neurons in any dimension require only up to
Unsupervised and Semi-Supervised Robust Spherical Space …
The results show that our method either surpasses or is competitive compared with the recent methods for both unsupervised and semi-supervised domain adaptation. Ablation studies also confirm the effectiveness of the spherical classifier, spherical discriminator, …
Ensembles of random sphere cover classifiers
The αRSC algorithm classifies a new case by the following rules:. 1. Rule 1. A test example that is covered by a sphere, takes the target class of the sphere. If there is more than one sphere of different target class covering the test example, the classifier takes the target class of the sphere with the closest centre.
A new maximal-margin spherical-structured multi-class
Support vector machines (SVMs), initially proposed for two-class classification problems, have been very successful in pattern recognition problems. For multi-class classification problems, the standard hyperplane-based SVMs are made by constructing and combining several maximal-margin hyperplanes, and each class of data is confined into a …
Nonparametric binary regression models with spherical
Spherical statistics is an analysis of data on a unit hyper-spherical domain. In this paper, we mainly consider the local kernel estimators for regression models with a binary response and the predictors including spherical variables. ... Compared with other classifiers, the models proposed in this paper have better classification performance ...
Robust Object Classification Approach using Spherical …
Robust Object Classification Approach using Spherical Harmonic. ectronic & Electrical Engineering, University College London, UKAbstract—In this paper, we present a robust …
Spherical Space Domain Adaptation With Robust Pseudo …
Adversarial domain adaptation (DA) has been an effective approach for learning domain-invariant features by adversarial training. In this paper, we propose a novel adversarial DA approach completely defined in spherical feature space, in which we define spherical classifier for label prediction and spherical domain discriminator for discriminating domain labels. To utilize …
Spherical Space Domain Adaptation With Robust Pseudo …
Adversarial domain adaptation (DA) has been an effective approach for learning domain-invariant features by adversarial training. In this paper, we propose a novel adversarial DA approach completely defined in spherical feature space, in which we define spherical classifier for label prediction and spherical domain discriminator for discriminating domain labels.
Twin pinball loss support vector hyper-sphere classifier for …
From the classification experiments for synthetic and UCI datasets, it can be clearly seen that TPSVH has better classification accuracy and generalization performance compared with other classifiers. Motivated by twin support vector hyper-sphere (TSVH) and support vector machine with pinball loss (pin-SVM), this paper formulated a twin pinball loss …
HDSS-Net: A Novel Hierarchically Designed Network With Spherical …
Finally, a novel spherical space classifier (SSC) is proposed to expand the interclass margin and compress the intraclass feature distribution by fully taking advantage of the property of spherical space. The experimental results on two benchmark datasets (OpenSARShip and FUSAR-Ship) jointly show that the proposed HDSS-Net performs better than ...
Unsupervised and Semi-Supervised Robust Spherical Space …
A novel adversarial domain adaptation approach defined in the spherical feature space, in which spherical classifier for label prediction and spherical domain discriminator for discriminating domain labels are defined and a spherical robust pseudo-label loss is developed to utilize pseudo-labels robustly. Adversarial domain adaptation has been an effective approach …
Robust Object Classification Approach using Spherical …
spherical convolution is applied on point clouds instead of spherical voxel grid, resulting in better rotation equivariant. In another work, a network called Deepsphere [8], spherical CNNs are used on graph represented shapes, where shapes are projected onto the sphere using HEALPix sampling, in which the relations between the pixels of the ...
CVPR 2020 Open Access Repository
Adversarial domain adaptation (DA) has been an effective approach for learning domain-invariant features by adversarial training. In this paper, we propose a novel adversarial DA approach …
XCS-based reinforcement learning algorithm for motion …
Results of several simulated experiments for the spherical robot show that this approach is capable of planning a near-optimal path to a predefined target from any given position/orientation. A Reinforcement Learning (RL) algorithm based on eXtended Classifier System (XCS) is used to navigate a spherical robot. Traditional motion planning strategies …
Spheronizer NEA|Sphere: Neuman & Esser
NEA|Sphere Solutions. Traditionally, the production of spherical anode graphite is based on the series connection of 20-30 impact classifier mills. The grinding, classifying and rounding processes are carried out simultaneously to achieve the particle size and distribution as well as the desired shape.
XCS-based reinforcement learning algorithm for motion …
A Reinforcement Learning (RL) algorithm based on eXtended Classifier System (XCS) is used to navigate a spherical robot. Traditional motion planning strategies rely on pre-planned optimal trajectories and feedback control techniques. The proposed learning agent approach enjoys a direct model-free methodology that enables the robot to function in dynamic …
Spherical Space Domain Adaptation With Robust Pseudo …
A novel adversarial DA approach completely defined in spherical feature space is proposed, in which spherical classifier for label prediction and spherical domain discriminator for discriminating domain labels are defined and a robust pseudo-label loss is developed. Adversarial domain adaptation (DA) has been an effective approach for learning domain-invariant features …
A Randomized Sphere Cover Classifier | SpringerLink
This paper describes an instance based classifier, the randomised sphere covering classifier (αRSC), that reduces the training data set size without loss of accuracy when compared to nearest neighbour classifiers.The motivation for developing this algorithm is the desire to have a non-deterministic, fast, instance based classifier that performs well in isolation but is also …
Spherical Transformer: Adapting Spherical Signal to …
Convolutional neural networks (CNNs) have been widely used in various vision tasks, e.g. image classification, semantic segmentation, etc. Unfortunately, standard 2D CNNs are not well suited for spherical signals such as panorama images or spherical projections, as the sphere is an unstructured grid.
A novel air-suction classifier for fresh sphere fruits in …
To improve the harvesting efficiency of the sphere fruits and reduce the workload of post-harvest classification, an air-suction sphere fruit classifier was designed in this paper, which can achieve the separation of high-quality fruit and common fruit through a pneumatic device according to the size and quality of the fruit when the cones are picked. First, three fruit …