Bayesian quadrature improves integration on Riemannian manifolds.
problem Efficiently computing integrals on nonlinear geometric data.
method Probabilistic numerical methods, specifically Bayesian quadrature, on Riemannian manifolds.
result Bayesian quadrature reduces the number of function evaluations compared to Monte Carlo methods.
Survey of Bayesian learning for neural networks.
problem Limitations of Bayesian learning in practical applications.
method Introduction to Bayesian Neural Networks and algorithms for inference.
result Discussion of standard and recent approaches for Bayesian inference in neural networks.
In-BO optimizes complex constrained domains using SIn-GP surrogate models.
problem Optimizing in complex constrained domains with irregular shapes.
method Sparse Intrinsic Gaussian Processes (SIn-GP) on manifolds with heat kernel estimation.
result In-BO outperforms traditional BO in complex constrained domains.
Bayesian hyperbolic MDS improves tree-like data representation.
problem Representing tree-like structures in high-dimensional data.
method Bayesian approach to hyperbolic MDS for low-dimensional manifold.
result Bayesian hyperbolic MDS reduces computational complexity and improves accuracy.
We reframe linear dimensionality reduction as a problem of Bayesian inference on matrix manifolds. This natural paradigm extends the Bayesian framework to dimensionality reduction tasks in higher dimensions with simpler models at greater speeds. Here an orthogonal basis is treated as a single point on a manifold and is…
New Bayesian matrix completion method using Stiefel manifolds.
problem Efficient Bayesian matrix completion with uncertainty quantification.
method Geodesic Hamiltonian Monte Carlo on Stiefel manifolds.
result Improved sampling performance and accuracy on real-world problems.
Bayesian imaging uses neural networks to learn prior knowledge from data.
problem Performing Bayesian inference in imaging problems with limited prior knowledge.
method Constructs a data-driven prior on a sub-manifold of the image space using neural networks, and performs Bayesian computation on this manifold.
result Established the existence and well-posedness of the posterior distribution and moments, and demonstrated superior performance compared to existing methods.
Bayesian analysis shows unlabeled data improve graph-based semi-supervised learning.
problem Improving semi-supervised learning with limited labeled data.
method Bayesian nonparametric approach using unlabeled data for graph-based learning.
result Posterior contracts optimally around the truth with sufficient unlabeled data.
Transformers mimic Bayesian reasoning in controlled settings, revealing geometric mechanisms.
problem Verifying if transformers perform Bayesian reasoning rigorously in natural data.
method Constructing Bayesian wind tunnels with known posteriors and proving memorization impossibility.
result Transformers achieve 10−3-10−4 bit accuracy in Bayesian posteriors, while MLPs fail. A new metric tensor improves Riemann manifold Monte Carlo for Bayesian models.
problem Improving sampling efficiency in Bayesian hierarchical models.
method Metric tensor derived from log-density gradient covariance matrices.
result Metric tensors enhance sampling for complex Bayesian models.
The paper reinterprets Bayesian priors and posteriors using Riemannian manifolds.
problem The dependence of maximum a posteriori estimates on parametrization.
method Assuming a Riemannian manifold with Fisher metric, the paper reinterprets priors and posteriors as distributions over probability distributions, making estimates independent of parametrization.
result A maximum a posteriori estimate independent of parametrization is defined.
This work investigates a mixture of LMC and RMHMC with MMALA for geometric ergodicity.
problem Lack of geometric ergodicity study in Riemannian manifold and Lagrangian Monte Carlo methods.
method Investigates a mixture of LMC and RMHMC with MMALA to achieve geometric ergodicity.
result Demonstrates geometric ergodicity in the mixture of LMC and RMHMC with MMALA.
Bayesian explanations are more resilient to adversarial attacks than deterministic ones.
problem Stability of saliency-based explanations under adversarial attacks in Neural Networks.
method Empirical and theoretical analysis of Bayesian vs deterministic Neural Networks.
result Bayesian explanations are more stable under adversarial perturbations and direct attacks.
Improves sampling efficiency for complex Bayesian models.
problem Inference challenges in hierarchical Gaussian-process models.
method Optimised Riemannian-manifold Hamiltonian Monte Carlo (RMHMC) with dynamic programming.
result Significant improvement in sampling efficiency and model evidence calculation.
A new Weyl prior is proposed for Bayesian statistics, offering a more canonical choice for parameter α.
problem Choosing a prior distribution for Bayesian inference.
method Proposed a new Weyl prior based on the Weyl structure on a statistical manifold.
result The Weyl prior is a special case of the α-parallel prior with α = -n, where n is the dimension of the statistical manifold.
Paper extends Bayesian Cramér-Rao bound with geometric considerations.
problem Estimation of covariance matrices with geometric structures.
method Intrinsic Bayesian Cramér-Rao bound with Riemannian geometry.
result Performance bounds for covariance matrix estimation.
This paper improves dependency networks using information geometry.
problem Technical disadvantage in dependency networks' learned distribution.
method Interpret pseudo-Gibbs sampling as iterative m-projections onto manifolds.
result Dependency networks can learn faster and have similar performance to Bayesian networks.
We review the information geometry of linear systems and its application to Bayesian inference, and the simplification available in the Kähler manifold case. We find conditions for the information geometry of linear systems to be Kähler, and the relation of the Kähler potential to information geometric quantities such …
This paper studies clustering for possibly high dimensional data (e.g. images, time series, gene expression data, and many other settings), and rephrase it as low rank matrix estimation in the PAC-Bayesian framework. Our approach leverages the well known Burer-Monteiro factorisation strategy from large scale optimisati…
New learning rule simplifies Bayesian updates for deep learning.
problem Bayesian learning rule's complexity and manifold constraints.
method Lie-group approach to simplify Bayesian updates.
result New algorithm learns sparse features in deep learning.
This paper tackles high-dimensional Bayesian optimization by projecting a manifold into a lower space.
problem High-dimensional optimization of expensive functions with limited labeled data.
method Random linear projection of a manifold embedded in high-dimensional space, combined with semi-supervised learning of the manifold's geometry.
result Our approach outperforms existing high-dimensional BO methods in various synthetic and real-world applications.
New tools for constructing disintegrations and studying their modes.
problem Difficulty in constructing disintegrations and understanding their modes.
method Developed comprehensive mathematical tools for constructing disintegrations and analyzing their modes.
result Disagreement between restricted density and disintegration density in certain cases.
Bayesian inverse problems use generative models for efficient inference.
problem Efficiently solving inverse problems with limited data and expert knowledge.
method Generative models trained on databases, Laplace approximation for prior density.
result Bayes estimates are consistent, not dependent on generative model quality.
A new method improves uncertainty quantification in Bayesian inference.
problem Poor uncertainty quantification in traditional Gibbs posteriors.
method Sequential Gibbs posteriors with a Bernstein-von Mises theorem.
result Sequential Gibbs posteriors provide better frequentist coverage.
We develop Riemannian Stein Variational Gradient Descent (RSVGD), a Bayesian inference method that generalizes Stein Variational Gradient Descent (SVGD) to Riemann manifold. The benefits are two-folds: (i) for inference tasks in Euclidean spaces, RSVGD has the advantage over SVGD of utilizing information geometry, and …
New analysis shows how cross-entropy training shapes attention in transformers.
problem Understanding how gradient-based learning creates the required internal geometry in transformers.
method Developed a first-order analysis of cross-entropy training effects on attention scores and values in a transformer attention head.
result Introduced an advantage-based routing law and responsibility-weighted update for attention scores and values, respectively.
This work explores how overparametrization and priors affect Bayesian neural network posteriors.
problem Symmetries, non-identifiabilities, and weight-space priors fragment and inflate BNN posteriors.
method We study the interplay between overparametrization and priors in BNN posteriors, deriving key phenomena and validating through experiments.
result Overparametrization induces structured, prior-aligned weight posterior distributions.
Deep Gaussian processes on manifolds improve performance on complex data.
problem Complex data on manifolds that shallow models struggle with.
method Residual deep Gaussian processes on Riemannian manifolds.
result Significant improvement in prediction quality and uncertainty calibration.
Paper accelerates Bayesian few-shot classification using mirror descent.
problem Non-conjugate inference in Bayesian few-shot classification.
method Integrates mirror descent-based variational inference into Gaussian process-based few-shot classification.
result Accelerated convergence and improved uncertainty quantification.
Bayesian framework reduces high-dimensional GP modeling costs.
problem Challenges in fitting Gaussian processes to high-dimensional inputs.
method Hierarchical Bayesian model with orthonormal projection matrix, incorporating Deep Gaussian Processes.
result Improves predictive performance and uncertainty quantification.
Algorithm improves variational inference in Wasserstein distance.
problem Improving variational inference methods for complex models.
method Wasserstein contraction analysis of coordinate ascent.
result General and sharp convergence guarantees for various models.
New analysis shows scores learn data manifolds better than distributions.
problem Learning the full distribution vs. just the data manifold.
method Novel analysis of scores in the small-σ regime.
result Scores learn data manifold information Θ(σ−2) stronger than distribution information. Injective flows for star-like manifolds improve variational inference efficiency.
problem Efficiently modeling densities on star-like manifolds with exact Jacobian computation.
method Proposed injective flows for star-like manifolds with exact Jacobian computation.
result Exact Jacobian computation for star-like manifolds reduces computational cost to NFs.
Bayesian method for knot inference in multivariate spline regression.
problem Inference on knot locations in multivariate spline regression due to non-differentiability and varying dimensions.
method Fully Bayesian approach with a new prior on knot number and analytic formula for normal model, extended Bayesian information criterion for non-normal cases, reversible jump Markov chain Monte Carlo.
result Demonstrated superior performance in function fitting with jumping discontinuity.
Let M be a smooth compact oriented manifold without boundary, imbedded in a euclidean space E and let f be a smooth map of M into a Riemannian manifold N. An unknown state x in M is observed via X=x+su where s>0 is a small parameter and u is a white Gaussian noise. For a given smooth prior on M and smooth estimators g …
Optimal persuasion involves projecting state vectors onto lower-dimensional 'optimal information manifolds'.
problem Optimal persuasion of another agent observing multi-dimensional data.
method Performing non-linear dimension reduction by projecting state vectors onto the 'optimal information manifold'.
result Optimal information design splits information into 'good' and 'bad' components, revealing only the direction of good information.
A new model clusters networks with community-specific submanifold structures.
problem Clustering networks with community-specific submanifold structures.
method Latent Structure Block Models (LSBM) for Bayesian spectral graph clustering.
result LSBM correctly recovers underlying communities in one-dimensional manifold structures.
Optimizes VAE hyperparameters for efficient training and manifold discovery.
problem Efficiently optimizing hyperparameters in VAEs for complex data.
method Latent Bayesian Optimization (zBO) for hyperparameter trajectory optimization.
result Demonstrated improved performance in finding joint rotationally invariant representations.
This paper examines linear embeddings for high-dimensional Bayesian optimization, identifying and addressing issues to improve performance.
problem Scaling Bayesian optimization to high-dimensional spaces while maintaining sample efficiency.
method Study and empirical evaluation of linear embeddings for BO, addressing design choices and their impact on performance.
result Properly addressing issues in linear embeddings significantly improves their efficacy in BO.
KPCA-BO improves BO for high-dimensional optimization problems by learning a non-linear sub-manifold.
problem High-dimensional optimization problems where Gaussian Process regression requires too much data and computation.
method KPCA-BO embeds a non-linear sub-manifold in the search space, learning a GPR model on this sub-manifold.
result KPCA-BO outperforms vanilla BO in convergence speed, especially as dimensionality increases.
Improved Bayesian inference using power priors with historical data.
problem Improving Bayesian inference with historical data.
method Generalized power priors that adapt to the α parameter of Amari's α-divergence. result Improved performance through appropriate choices of the α parameter. We consider the problem of recovering a function input of a differential equation formulated on an unknown domain M. We assume to have access to a discrete domain Mn={x1,…,xn}⊂M, and to noisy measurements of the output solution at p≤n of those points. We introduce a graph-based Bayesian inve…
Gaussian variational approximation is a popular methodology to approximate posterior distributions in Bayesian inference especially in high dimensional and large data settings. To control the computational cost while being able to capture the correlations among the variables, the low rank plus diagonal structure was in…
Bayesian framework for SSP problem learns optimal strategy through interactions.
problem Sequential decision-making in stochastic shortest path problems.
method Develops a Bayesian framework to learn optimal action-value function Q∗ through interactions, avoiding unrealistic assumptions. result Demonstrates data efficiency and uncertainty quantification compared to other methods.
Novel methods improve Bayesian analysis of chaotic dynamical systems.
problem Bayesian parameter inference and trajectory reconstruction of chaotic systems with sparse and noisy data.
method Pilot MAGI (pMAGI) and Pilot MAGI Sequential Prediction (PMSP) methods.
result pMAGI and PMSP significantly outperform existing methods in accuracy and computational efficiency.
Bayesian parametric matrix models provide uncertainty quantification for spectral learning.
problem Uncertainty quantification in spectral learning for safety-critical applications.
method Bayesian parametric matrix models (B-PMMs) that extend PMMs to provide uncertainty estimates.
result B-PMMs achieve exceptional uncertainty calibration (ECE < 0.05) while maintaining favorable scaling.
Bayesian neural networks improve uncertainty quantification in non-linear dimensionality reduction.
problem Current neural network models lack adequate uncertainty quantification.
method Deploy Markov chain Monte Carlo sampling algorithms for Bayesian inference in ANN models with latent variables.
result New research directions are needed due to fundamental challenges in neural networks with latent variables.
Bayesian neural networks can be simplified by parameterizing weights as rank-r matrices, reducing parameter count and improving performance.
problem High parameter count in standard Bayesian neural networks.
method Parameterize weights as W=ABop with A∈Rmimesr, B∈Rnimesr, inducing a singular posterior. result PAC-Bayes generalization bounds and loss bounds show improved performance with fewer parameters.