expectation–maximization algorithm
PulseAugur coverage of expectation–maximization algorithm — every cluster mentioning expectation–maximization algorithm across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New research details EM and MoM for Softmax Mixture Models
A new research paper explores the application of the Expectation-Maximization (EM) algorithm and the Method of Moments (MoM) to Softmax Mixture Models (SMMs). These models are used for analyzing probabilities in heterog…
-
Score Matching Linked to ML and EM in Mixed Linear Regression
Researchers have established a theoretical connection between score matching, maximum likelihood estimation, and the expectation-maximization (EM) algorithm within the context of mixed linear regression models. Their an…
-
New methods enhance Mixture-of-Experts model efficiency and performance
Researchers have developed new methods to improve the efficiency and performance of Mixture-of-Experts (MoE) models. One approach, Layer-wise Distribution Alignment (LDA), addresses the performance degradation that occu…
-
New EM algorithm improves handling of missing data in matrix-variate models
Researchers have developed an efficient expectation-maximization (EM) algorithm designed to handle missing data in matrix-variate normal mixture models. This new algorithm significantly reduces computational costs assoc…
-
New CP-GAMP algorithm accelerates Bayesian tensor reconstruction
Researchers have developed a new algorithm called CP generalized approximate message passing (CP-GAMP) for Bayesian CANDECOMP/PARAFAC (CP) decomposition. This method significantly reduces computation time compared to tr…
-
New robust clustering model uses Gaussian-Cauchy mixtures for outlier detection
Researchers have developed a new model-based clustering technique that utilizes mixtures of multivariate pseudo-Voigt distributions. This approach combines Gaussian and Cauchy distributions to enhance robustness in clus…
-
New framework enhances interpretable machine learning with heterogeneous experts
Researchers have developed a new framework for interpretable machine learning by extending the Mixture of Experts (MoE) model. This novel approach allows for heterogeneous experts, incorporating decision trees, linear s…
-
Gradient EM algorithm achieves global convergence for over-parameterized Gaussian Mixtures
Researchers have established a global convergence guarantee for the gradient Expectation-Maximization (EM) algorithm when applied to over-parameterized Gaussian Mixture Models (GMMs). This marks the first such result fo…
-
New EMS Coreset algorithm offers efficient data subsetting for machine learning
Researchers have developed EMS Coreset, a novel algorithm designed to create representative data subsets for machine learning tasks more efficiently. This method utilizes an expectation-maximization approach with Sinkho…
-
New bandit algorithm tackles LLM refinement with reward decay modeling
Researchers have developed a new contextual bandit algorithm designed to improve iterative refinement in Large Language Models (LLMs). This algorithm explicitly models reward decay, addressing the issue of over-exploita…
-
New FLAMECHE method enhances privacy in Clustered Federated Learning
Researchers have introduced FLAMECHE, a novel approach to Clustered Federated Learning (CFL) that addresses the inherent trade-offs between privacy, communication cost, and computational efficiency. FLAMECHE reformulate…
-
New CW-EDMD method improves Koopman operator approximation for complex systems · 2 sources tracked
Researchers have developed Cluster-Weighted Extended Dynamic Mode Decomposition (CW-EDMD), a novel method for approximating Koopman operators from data. This approach addresses the inefficiency of single global operator…
-
New MSFA framework tackles high-dimensional spatial data clustering
Researchers have introduced a novel mixture of spatial factor analyzers (MSFA) designed to tackle the complexities of clustering high-dimensional spatial data. This framework utilizes a spline-based spatial decay covari…
-
New Vision Transformer Cuts Image Captioning Costs with Clustering
Researchers have developed a new vision transformer architecture that significantly reduces computational costs for image captioning. By replacing the standard self-attention mechanism with a Gaussian Mixture Model-base…
-
New EM-NeSy approach enhances neurosymbolic AI learning
Researchers have introduced EM-NeSy, a novel approach to neurosymbolic learning that frames the process as an instance of the Expectation-Maximization (EM) algorithm. This method allows for approximate inference without…
-
New BI-BAU Method Aims for Complete Backdoor Unlearning in AI Models
Researchers have proposed a new method called Blind Inversion-Backdoor Adversarial Unlearning (BI-BAU) to address the limitations of current backdoor defenses in AI models. This approach frames backdoor unlearning as a …
-
New Kalman Filter framework models complex time-series data on cell complexes
Researchers have developed a new topology-aware state space framework for inferring latent dynamics from complex time-series data. This approach utilizes stochastic partial differential equations on cell complexes to mo…
-
New entropic optimal transport loss improves model-based clustering methodology
Researchers have developed a novel loss function for model-based clustering using entropic optimal transport. This new approach aims to overcome the limitations of traditional log-likelihood optimization, which can suff…
-
Physically-informed fuzzy clustering method separates ionogram tracks
Researchers have developed a new physically-informed fuzzy clustering method to analyze vertical sounding ionograms. This technique automatically separates ionograms into distinct tracks, even in disturbed ionospheric c…
-
New HPPCA model improves analysis of longitudinal data with missing values
Researchers have developed Hierarchical Probabilistic Principal Component Analysis (HPPCA), a novel statistical model designed to handle complex longitudinal data with missing values. This two-level probabilistic factor…