Wasserstein metric
PulseAugur coverage of Wasserstein metric — every cluster mentioning Wasserstein metric across labs, papers, and developer communities, ranked by signal.
- instance of ScienceCast 90%
- instance of Gotit.pub 90%
- instance of optimal transport 90%
- used by Gaussian function 90%
- used by Gotit.pub 70%
- instance of alphaXiv 70%
- instance of DagsHub 70%
- used by CatalyzeX 70%
- instance of CatalyzeX 70%
- used by optimal transport 70%
- instance of Gaussian function 70%
- instance of Sliced-Wasserstein distance 70%
10 day(s) with sentiment data
-
New research details Wasserstein distance for parameter estimation complexity
This paper explores the complexity of estimating translation and shrinkage parameters for distributions, a fundamental challenge in machine learning. While Maximum Likelihood Estimation (MLE) for this problem is NP-hard…
-
Researchers Quantify Transformer Concentration Phenomena in LLMs
A new research paper explores the concentration phenomena of transformers used in large language and foundation models. The study, focusing on encoder-only transformers at inference time, describes their behavior using …
-
New Riemannian geometry for Sinkhorn divergence established
Researchers have developed a new framework for understanding the geometry of Sinkhorn divergence, a widely used measure in optimal transport. This work establishes a formal Riemannian structure and derives key geometric…
-
Quantum Evolution Kernel for Node-Attributed Graphs Unveiled
Researchers have developed a new method to embed node-attributed graphs onto neutral-atom quantum computers, extending PASQAL's quantum evolution kernel (QEK). This approach uses a node-level feature encoding based on t…
-
New method generates differentially private synthetic data using moment matching
Researchers have developed a new method for generating differentially private synthetic data using generalized moment matching. This technique estimates moments privately, fits a probability measure, and then samples sy…
-
New estimator advances quantile distributional reinforcement learning accuracy
Researchers have developed a new estimator for quantile distributional reinforcement learning, aiming to accurately approximate the return distribution under a given policy. The study establishes non-asymptotic error bo…
-
New TPO framework enhances LLM preference alignment against distribution shifts
Researchers have introduced TPO, a novel framework designed to address distribution shift challenges in Direct Preference Optimization (DPO) for large language models. TPO tackles the hierarchical nature of preference d…
-
NeurPRISE model enhances scenario reduction for robust optimization
Researchers have developed NeurPRISE, a novel neural surrogate model designed to improve the efficiency of scenario reduction in two-stage robust optimization problems with discrete uncertainty. This model, built on a G…
-
New Sliced Wasserstein Sketches Offer Compact Probability Distribution Representations
Researchers have developed a method called Sliced Wasserstein Sketches to create compact representations of probability distributions. These sketches allow for efficient and geometrically meaningful comparisons, reducin…
-
New method improves Gaussian approximation for wide neural networks
Researchers have developed a method to improve the convergence rate of Gaussian approximations for wide neural networks. This new technique lifts finite-sample bounds to functional bounds, achieving a convergence rate o…
-
New method improves signal recovery in one-bit compressed sensing
Researchers have developed a new method for recovering signals from one-bit compressed sensing using posterior sampling. This approach achieves accurate recovery with high probability when the number of measurements sca…
-
New mathematical framework identifies low-rank measures from projections
A new mathematical framework called the Silhouette Operator has been developed to reconstruct high-dimensional objects from low-dimensional projections. This operator is designed for low-rank signed measures on R^2, whi…
-
New technique analyzes Q-learning convergence and bias in stochastic approximation
Researchers have developed a new technique for analyzing nonsmooth contractive stochastic approximation (SA) dynamics, particularly relevant to Q-learning. The study establishes weak convergence of iterates to a station…
-
New Gaussian Approximation Bounds for Markov Chains Developed
Researchers have developed new Gaussian approximation bounds for sums of multivariate martingale differences derived from uniformly ergodic Markov chains. These bounds, expressed in higher-order Wasserstein distance, ac…
-
Quantum algorithms promise speedups for sampling and optimization
Researchers have developed new quantum algorithms that offer speedups for sampling from complex probability distributions and for non-convex optimization tasks. These algorithms enhance classical methods like Langevin M…
-
Bayesian Experimental Design: KL Divergence vs. Wasserstein Distance
A new paper published on arXiv explores the use of Bayesian experimental design (BED) for calibrating model discrepancies. The research compares Kullback-Leibler (KL) divergence and Wasserstein distance as utility funct…
-
AI models in pathology gain robustness through new benchmarking and artifact generation techniques
Two new research papers explore methods for improving the reliability and robustness of AI models in computational pathology. The first paper, "Reliable Benchmarking of Artifact Detection in Computational Pathology," pr…
-
Review paper details optimal transport for network comparison in ML
A new review paper explores the application of optimal transport methods for comparing networks, particularly in machine learning contexts. The paper details three primary distances: Wasserstein, Gromov-Wasserstein, and…
-
New Bayesian online learning framework preserves fast regret with approximations
Researchers have developed a new framework for Bayesian online learning that preserves fast predictive regret guarantees even with approximate posterior computations. The study demonstrates that the accuracy of the appr…
-
New Fréchet regression method uses nonparanormal transport for multivariate distributions
Researchers have developed a new regression approach for multivariate distributional responses, which models distributions within the semiparametric nonparanormal family. This method incorporates the nonparanormal trans…