Related Experiment Video
Updated: Jul 12, 2026
07:05
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
Published on: October 27, 2016
Provable cluster-preserving visualizations with curvature-based stochastic neighbor embeddings
Tristan Luca Saidi1, Abigail Hickok2, Bastian Rieck3
1Department of Computer Science, Columbia University, New York, NY 10027.
Summary
EmbedOR, a new Stochastic Neighbor Embedding (SNE) algorithm, improves data visualization by incorporating discrete graph curvature. This method better preserves data geometry and cluster structures compared to UMAP and tSNE.
Area of Science:
- Data Visualization
- Machine Learning
- Computational Geometry
Background:
- Stochastic Neighbor Embedding (SNE) algorithms like UMAP and tSNE often fail to preserve the geometry of high-dimensional data.
- These methods can spuriously separate connected components and miss clusters in complex datasets.
Purpose of the Study:
- To introduce EmbedOR, a novel SNE algorithm designed to overcome the limitations of existing methods.
- To enhance data visualization by preserving geometric structures and improving cluster identification.
Main Methods:
- EmbedOR utilizes a curvature-enhanced distance metric within its stochastic embedding process.
- The algorithm incorporates discrete graph curvature to emphasize underlying cluster structures.
- Theoretical consistency results for tSNE are extended to a broader class of datasets using the EmbedOR metric.
Main Results:
- Extensive experiments on synthetic and real data demonstrate EmbedOR's superior visualization and geometry-preservation capabilities.
- EmbedOR significantly reduces the fragmentation of continuous, high-density data regions compared to UMAP and tSNE.
- The EmbedOR distance metric proves effective in annotating existing visualizations to identify fragmentation.
Conclusions:
- EmbedOR offers a robust solution for visualizing high-dimensional data, addressing key limitations of current SNE techniques.
- The incorporation of graph curvature enhances the preservation of underlying data geometry and cluster integrity.
- EmbedOR provides a valuable tool for deeper insights into data structure and for annotating complex visualizations.
Related Concept Videos
01:25Curvature and Its Interpretation
Curvature describes how rapidly a curve changes direction at a particular point. A curve with a small curvature bends gently, while a curve with a large curvature turns sharply. For a space curve, the position of a moving object can be described by a vector-valued function r(t), where t often represents time. The direction of motion is determined by the tangent vector, and the unit tangent vector is obtained by normalizing the derivative of the position vector.The unit tangent vector gives the...
01:20Divergence Theorem in 3D Space
In vector calculus, flux measures the total flow of a vector field through a surface. For a closed surface in three-dimensional space, this means measuring how much of the field passes outward through every point on the boundary. Directly calculating this flux can be difficult when the surface has a complicated or irregular shape. The Divergence Theorem provides a powerful alternative by relating surface flux to behavior inside the enclosed region.The Divergence Theorem states that the outward...
01:20Cluster Sampling Method
Appropriate sampling methods ensure that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
01:29Conservative Vector Fields
A conservative vector field describes a force or field in which the work done between two points depends only on the initial and final positions. For a ball moving in Earth’s gravitational field, gravity performs work determined by the difference in height, regardless of whether the ball moves vertically or follows a curved trajectory.A vector field is conservative if it can be expressed as the gradient of a scalar potential function, f. In two dimensions, this is written...
01:06Divergence and Stokes' Theorems
The divergence and Stokes' theorems are a variation of Green's theorem in a higher dimension. They are also a generalization of the fundamental theorem of calculus. The divergence theorem and Stokes' theorem are in a way similar to each other; The divergence theorem relates to the dot product of a vector, while Stokes' theorem relates to the curl of a vector. Many applications in physics and engineering make use of the divergence and Stokes' theorems, enabling us to write numerous physical laws...
01:19Degree of Curvature and Radius of Curvature
The degree of curvature and the radius of curvature are fundamental concepts in determining the sharpness or smoothness of a curve. The degree of curvature is a measure of how steeply a curve bends and can be determined using the chord basis or the arc basis. In the chord basis method, the degree of curvature is defined as the central angle subtended by a chord of 30.48 meters, helping in the calculation of the radius of the curve. The arc basis method defines the degree of curvature as the...