Markov chain Monte Carlo
PulseAugur coverage of Markov chain Monte Carlo — every cluster mentioning Markov chain Monte Carlo across labs, papers, and developer communities, ranked by signal.
12 day(s) with sentiment data
-
New Bayesian Framework Enhances Feature Extraction for Spatio-Temporal Data
Researchers have developed a new Bayesian feature extraction framework designed for high-dimensional spatio-temporal data, particularly useful in scientific domains. This framework utilizes Gaussian and Diffused-gamma p…
-
New neural sampler ATLAS revolutionizes amorphous material research
Researchers have developed ATLAS, a novel foundation neural sampler designed to efficiently generate and study amorphous materials. This model utilizes a diffusion process learned by an equivariant graph neural network …
-
New algorithm boosts Neural Markov Logic Networks for relational structure generation
Researchers have introduced Parallel Noising, a novel training and inference algorithm designed to enhance Neural Markov Logic Networks (NMLNs). This new method, inspired by parallel-tempering Markov chain Monte Carlo t…
-
New neural diffusion method enables efficient spatial simulation
Researchers have introduced Neural Conditional Simulation (NCS), a novel method for simulating spatial processes. NCS utilizes neural diffusion models to generate samples from predictive distributions, which are often i…
-
BayesPO framework uses Bayesian sampling for LLM prompt optimization
Researchers have developed BayesPO, a novel framework for optimizing prompts in large language models without altering model parameters. This method treats prompt optimization as Bayesian posterior sampling, combining t…
-
New Bayesian inference method uses energy distance for faster sampling
Researchers have developed a new method for amortized Bayesian inference, particularly useful for nonlinear inverse problems. This technique learns a reusable map that can quickly generate posterior samples for new obse…
-
New active learning strategy improves bioacoustic classification for rare calls
Researchers have developed a new active learning strategy called BADGE-Greedy-DPP for bioacoustic call-type classification, which is particularly effective for long-tailed and sparse datasets. This method greedily selec…
-
Quantum-classical sampling methods compared to classical MCMC
A new research paper explores a hybrid quantum-classical approach for sampling discrete Markov random fields, a computationally challenging task. The study compares quantum sampling methods against classical Markov Chai…
-
SalientGS unifies SfM and 3DGS for faster 3D scene reconstruction · 2 sources tracked
Researchers have developed SalientGS, a novel pipeline that unifies Structure-from-Motion (SfM) with 3D Gaussian Splatting (3DGS) for 3D scene reconstruction. The system employs importance-guided Markov Chain Monte Carl…
-
New sampling methods improve efficiency for complex distributions · 2 sources tracked
Researchers have developed a new method called Gradient-free Riemannian Langevin Sampler (GRiLS) to improve the efficiency of sampling multimodal probability distributions. This approach aims to overcome limitations in …
-
New framework enhances neural likelihood approximation for complex Bayesian problems
Researchers have developed a new framework for neural likelihood approximation in Bayesian inverse problems, addressing challenges posed by complex scientific and engineering models. This approach trains likelihood surr…
-
PRISM3D framework reconstructs 3D scenes from extreme motion blur
Researchers have developed PRISM3D, a novel framework for 3D scene reconstruction from severely motion-blurred images, a task where traditional methods fail. The system employs a Robust Initialization strategy using dee…
-
New Bayesian GLMMs integrate neural encoders for multimodal data analysis
Researchers have developed a novel method to integrate neural encoders into Bayesian Generalized Linear Mixed Models (GLMMs). This approach allows GLMMs to handle high-dimensional data modalities like images and text, w…
-
New approach quantifies neural network uncertainty using gradient norms
Researchers have developed a novel method for quantifying uncertainty in neural networks, particularly large language models, by approximating predictive uncertainty using gradient norms and an isotropy assumption. This…
-
Diffusion models accelerate thermalization in condensed matter physics simulations
Researchers have developed a new diffusion model technique for efficiently sampling spin-system states with continuous symmetries, specifically applied to the XY model in condensed matter physics. This method overcomes …
-
Diffusion models accelerate Schwinger model sampling in physics research
Researchers have explored a novel diffusion-based method for accelerating the sampling of the Schwinger model, a problem in lattice quantum field theory. They developed a U(1)-equivariant score-based generative model to…
-
New Gibbs distribution enhances Monte Carlo integration accuracy
Researchers have developed a novel Gibbs distribution designed to improve Monte Carlo integration methods. This distribution's support concentrates around MMD minimizers as a temperature parameter decreases, offering ti…
-
Simulation-based inference offers faster Bayesian calibration for epidemiological models
A new research paper proposes simulation-based inference (SBI) as a faster and more efficient alternative to Markov chain Monte Carlo (MCMC) for calibrating epidemiological models. The study, which used COVID-19 ICU occ…
-
New Laplace--Fisher Gate Identity Enhances Score Estimation in Bayesian Inverse Problems
Researchers have developed a new method called the Laplace--Fisher Gate Identity (LFGI) for estimating scores in sampling from unnormalized targets. This method uses matrix-valued blending coefficients, or gates, to opt…
-
New workflow synergizes MCMC and Gaussian Processes for chemical reaction discovery
Researchers have developed a novel gray-box workflow called PC-MCMC-CIGP that integrates physically constrained Markov Chain Monte Carlo (MCMC) sampling with Chemical-Informed Gaussian Processes (CIGP) for discovering r…