t-Distributed Stochastic Neighbor Embedding
PulseAugur coverage of t-Distributed Stochastic Neighbor Embedding — every cluster mentioning t-Distributed Stochastic Neighbor Embedding across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New AI model uses sparse routing for retinal pathology analysis
Researchers have developed a new deep learning architecture for analyzing retinal fundus images that utilizes sparse conditional computation. This model pairs a Guided Context Gating (GCG) spatial attention front-end wi…
-
Social media analysis reveals tight right-wing clusters, dispersed left-wing content
An analysis of social media platforms reveals distinct clustering patterns based on user-shared domain names. The study found that right-wing content on Twitter forms a tightly knit cluster, leading to predictable recom…
-
New FloDR method offers invertible dimensionality reduction with normalizing flows
Researchers have introduced FloDR, a novel invertible dimensionality reduction method that utilizes a normalizing flow. Unlike traditional methods like t-SNE and UMAP, which discard information during the optimization p…
-
Paper warns of widespread misuse of t-SNE and UMAP in visual analytics
A new paper published on arXiv highlights the widespread misuse of dimensionality reduction techniques like t-SNE and UMAP in visual analytics. The research indicates that practitioners often misinterpret these tools, u…
-
New modular approach enhances data visualization transparency
Researchers have developed a new modular approach for data visualization that first clusters the data, then embeds each cluster individually, and finally aligns these clusters to create a global embedding. This method a…
-
New ML method NESS improves single-cell data analysis
Researchers have developed NESS, a new machine learning approach designed to improve the representation of single-cell data. This method builds upon the Predictability-Computability-Stability (PCS) framework to address …
-
Unsupervised AI models can learn sensitive attributes, violating fairness
Researchers have demonstrated that unsupervised machine learning representations can inadvertently encode sensitive attributes like age and income, even when these attributes are excluded from the training data. A new m…
-
EntroPath: New manifold learning method uses path ensembles
Researchers have introduced EntroPath, a novel manifold learning method designed to reconstruct geodesic geometry from data graphs. This method utilizes ensembles of diffusion paths, specifically employing a maximum ent…
-
New WIDER-FAIR dataset reveals bias in face detection models
Researchers have introduced WIDER-FAIR, a new dataset designed to evaluate fairness in face detection models. Built upon the WIDER-FACE benchmark, WIDER-FAIR includes manual annotations for perceived ethnicity and sex a…
-
New methods enhance representation learning with improved interpretability
Researchers have developed new dimensionality reduction methods that go beyond optimizing variance or correlation to improve statistical dependence, data diversity, contrast, and interpretability. These methods combine …
-
New RGNet architecture tackles class imbalance in fault diagnosis
Researchers have developed RGNet, a novel neural network architecture inspired by the renormalization group (RG) concept, designed to address challenges like class imbalance and multidimensional noise in machine learnin…
-
AI improves MR reconstruction generalization for neonatal imaging
Researchers have developed new methods to improve the generalization of deep learning models for MR reconstruction, specifically for adult-to-neonatal brain imaging. By employing contrast-informed data augmentation and …
-
New IRIS algorithm visualizes time-structured biomedical data
Researchers have developed IRIS, a novel manifold learning algorithm designed to visualize high-dimensional biomedical data that changes over time. Unlike existing methods, IRIS can structure its layouts chronologically…
-
MEDAL framework enables quantitative validation of manifold embeddings
Researchers have introduced MEDAL (Manifold Embedding Distillation via Autoencoder Learning), a new framework designed to quantitatively validate manifold embeddings. MEDAL distills existing embeddings into an encoder-d…
-
PCA visualization limitations highlighted with fossil teeth data
Researchers have identified limitations in Principal Component Analysis (PCA) when applied to visualizing high-dimensional data that resides on a nonlinear manifold. Using a dataset of fossil teeth, they demonstrated th…
-
New tool ParamInter visualizes high-dimensional parameter spaces for optimization
Researchers have developed a new tool called ParamInter designed to analyze high-dimensional input parameter spaces. This tool facilitates exploration of interpolations towards optimal parameter sets using guided analyt…
-
DR-SNE enhances dimensionality reduction by preserving data density
Researchers have introduced DR-SNE, a new dimensionality reduction technique that addresses distortions in data density often seen with methods like t-SNE. DR-SNE reformulates the process to jointly align conditional st…
-
UMAP dimensionality reduction method compared to PCA and t-SNE
A new paper compares Uniform Manifold Approximation and Projection (UMAP) with other dimensionality reduction techniques like PCA and t-SNE. The study systematically evaluates supervised UMAP for both regression and cla…
-
New Class Angular Distortion Index metric improves dimensionality reduction faithfulness
Researchers have introduced the Class Angular Distortion Index (CADI), a novel metric for evaluating dimensionality reduction techniques. CADI addresses limitations in existing metrics by assessing the faithfulness of c…
-
VERA tool automatically explains 2D data embeddings with region annotations
Researchers have developed VERA, a new method for automatically generating visual explanations of two-dimensional data embeddings. VERA identifies key regions within these embeddings and links them to human-interpretabl…