Bayesian inference
PulseAugur coverage of Bayesian inference — every cluster mentioning Bayesian inference across labs, papers, and developer communities, ranked by signal.
6 day(s) with sentiment data
-
New method enables scalable Bayesian inference for large neural networks
Researchers have developed Compressed Active Subspaces (CAS), a novel method to make Bayesian inference more scalable for large models. Traditional active subspace methods require significant memory for model gradients,…
-
New Bayesian framework enhances metabolite quantification in MRS
Researchers have developed a new Bayesian inference framework utilizing Sylvester normalizing flows (SNFs) to improve metabolite quantification in magnetic resonance spectroscopy (MRS). This physics-informed approach in…
-
New paper links Turing machines to singularities in analytic functions
A new paper by William Troiani explores a novel correspondence between the structure of Turing machines and the singularities of real analytic functions. This connection is established by linking linear logic's Ehrhard-…
-
New Bayesian inference method uses thermodynamic cycles to measure non-Gaussianity
Researchers have introduced the concept of Markov chain Monte Carlo (MCMC) cycles, drawing an analogy to thermodynamic cyclic processes in heat engines, to analyze Bayesian inference problems. They developed adaptive en…
-
New arXiv Paper Unifies Probability as Model Predictions
A new paper on arXiv proposes a unifying perspective on probabilities, arguing that all probabilities are outputs of prediction methods. This viewpoint suggests that probabilities are inherently model-dependent, even th…
-
AI framework enables efficient partial inverse design for high-performance concrete
Researchers have developed a novel cooperative neural network (CoNN) framework to address the complex challenge of partial inverse design for high-performance concrete (HPC). This AI-driven approach integrates an imputa…
-
New active inference method guides autonomous reconnaissance agents
Researchers have developed a novel active inference method for planning the routes of intelligent agents in autonomous reconnaissance missions. This approach aims to maintain a common operational picture by constructing…
-
Game Theory Enhances Drone Swarm Defense Capabilities
Researchers have explored the application of differential game theory to enhance drone swarm defense strategies. This approach models the opposing swarm as a rational agent, aiming to find a Nash equilibrium between def…
-
New HyperMC framework optimizes SGMCMC hyperparameters
Researchers have developed HyperMC, a novel framework for optimizing hyperparameters in Stochastic Gradient Markov Chain Monte Carlo (SGMCMC) methods. This approach utilizes a multi-fidelity tuning strategy, combining H…
-
Hamiltonian Monte Carlo explained from a probabilistic perspective
This article delves into Hamiltonian Monte Carlo (HMC), a sophisticated algorithm that powers modern Bayesian inference and statistical machine learning frameworks like PyMC. While originating from physics, HMC is a var…
-
New Bayesian control framework integrates spike-based neural models
Researchers have developed a novel Bayesian control framework that merges spike-based dynamics with probabilistic inference for adaptive control. This framework utilizes a biologically inspired spiking neural model comb…
-
New BTYD models leverage amortized variational inference for faster customer analysis
Researchers have developed a new family of Buy-'Til-You-Die (BTYD) models that move beyond the traditional Poisson process assumption. These new models utilize a Weibull renewal process and employ an amortized variation…
-
New research frames MDP planning as Bayesian inference over policies
Researchers have proposed a novel approach to Markov decision process (MDP) planning by framing it as a problem of Bayesian inference over policies. This conceptual shift treats the policy itself as a latent variable, w…
-
New framework for proportional analogies in probability distributions introduced
Researchers have introduced a new framework for proportional analogies applied to probability distributions, utilizing Bayesian updating as the core mechanism. This approach defines analogies as transformations between …
-
Generative Models Enhance Monte Carlo Sampling Techniques · 2 papers
Two recent arXiv papers explore the use of generative models to enhance sampling techniques in complex probability distributions. The first paper introduces a generator-guided inverse sampling method for Lévy-driven gen…
-
New research compares MCMC, LA, and VI complexity for generalized linear models
A new arXiv paper explores the computational complexity of Markov Chain Monte Carlo (MCMC) methods for generalized linear models, comparing them to Laplace approximation (LA) and variational inference (VI). The research…
-
New EPIK approach enhances Bayesian learning for software verification
Researchers have developed EPIK, a novel approach that integrates Bayesian learning with quantitative verification to analyze software system properties like reliability and response time. EPIK addresses the challenge o…
-
New adaptive technique reconstructs bosonic quantum states efficiently
Researchers have developed an adaptive reconstruction technique to more efficiently characterize bosonic quantum states. This method uses Bayesian inference, bootstrap, and active learning to select optimal measurement …
-
New theory defines and measures "forgetting" in machine learning algorithms
Researchers have proposed a new theoretical framework to understand and quantify "forgetting" in machine learning algorithms. This theory defines forgetting as a lack of self-consistency in a learner's predictive distri…
-
New Deep Sigma-Point Process Enhances SAR Imagery RCS Modeling
Researchers have developed a Deep Sigma-Point Process (DSPP) model to improve radar cross-section (RCS) modeling for spaceborne synthetic aperture radar (SAR) imagery. This new model utilizes a hierarchical Gaussian pro…