arXiv 2202.105222022-02-21cs.LGcs.AR

Non-Volatile Memory Accelerated Posterior Estimation

Boston University（波士顿大学）；IBM Research（IBM研究院）；Hasso Plattner Institute（哈索·普拉特纳研究所）

Andrew Wood, Moshik Hershcovitch, Daniel Waddington, Sarel Cohen, Peter Chin

英文摘要：

Bayesian inference allows machine learning models to express uncertainty. Current machine learning models use only a single learnable parameter combination when making predictions, and as a result are highly overconfident when their predictions are wrong. To use more learnable parameter combinations efficiently, these samples must be drawn from the posterior distribution. Unfortunately computing the posterior directly is infeasible, so often researchers approximate it with a well known distribution such as a Gaussian. In this paper, we show that through the use of high-capacity persistent storage, models whose posterior distribution was too big to approximate are now feasible, leading to improved predictions in downstream tasks.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Non-Volatile Memory Accelerated Geometric Multi-Scale Resolution Analysis

IBM Research（IBM研究院）；Hasso Plattner Institute（哈索·普拉特纳研究所）；Williams College（威廉姆斯学院）；Seoul National University（首尔国立大学）

Dimensionality reduction algorithms are standard tools in a researcher's toolbox. Dimensionality reduction algorithms are frequently used to augment downstream tasks such as machine learning, data science, and also are exploratory methods for understanding complex phenomena. For instance, dimensionality reduction is commonly used in Biology as well as Neuroscience to understand data collected from biological subjects. However, dimensionality reduction techniques are limited by the von-Neumann architectures that they execute on. Specifically, data intensive algorithms such as dimensionality red…

2022-02-21cs.ARcs.LG

Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling

University of Cambridge（剑桥大学）；Google Research（谷歌研究院）

We introduce repriorisation, a data-dependent reparameterisation which transforms a Bayesian neural network (BNN) posterior to a distribution whose KL divergence to the BNN prior vanishes as layer widths grow. The repriorisation map acts directly on parameters, and its analytic simplicity complements the known neural network Gaussian process (NNGP) behaviour of wide BNNs in function space. Exploiting the repriorisation, we develop a Markov chain Monte Carlo (MCMC) posterior sampling algorithm which mixes faster the wider the BNN. This contrasts with the typically poor performance of MCMC in hi…

2022-06-15cs.LGstat.ML

动力学朗之万与分裂吉布斯相遇：基于扩散先验的成像逆问题加速后验采样

Kinetic Langevin Meets Split Gibbs: Accelerated Posterior Sampling for Imaging Inverse Problems with Diffusion Priors

Hokkaido University（北海道大学）；Institute of Information Technology, VAST（越南科学技术翰林院信息技术学院）

本文提出RED-KLwSGS及其联合变体，将动力学朗之万扩散与分裂吉布斯采样结合，以加速成像逆问题的后验采样，并证明非渐近收敛性，实验显示更快收敛和高质量重建。

2026-10-08cs.LGstat.ML

Parameter Learning and Change Detection Using a Particle Filter With Accelerated Adaptation

University of Technology Sydney（悉尼科技大学）

This paper presents the construction of a particle filter, which incorporates elements inspired by genetic algorithms, in order to achieve accelerated adaptation of the estimated posterior distribution to changes in model parameters. Specifically, the filter is designed for the situation where the subsequent data in online sequential filtering does not match the model posterior filtered based on data up to a current point in time. The examples considered encompass parameter regime shifts and stochastic volatility. The filter adapts to regime shifts extremely rapidly and delivers a clear heuris…

2018-06-14cs.CEcs.LGcs.NE

用于旋转圆盘系统时空热通量贝叶斯推断的深度代理加速延迟接受HMC

Deep surrogate accelerated delayed-acceptance HMC for Bayesian inference of spatio-temporal heat fluxes in rotating disc systems

University of Bath（巴斯大学）

本文提出神经网络代理与延迟接受HMC结合的贝叶斯反演方法，自适应推断旋转圆盘系统的时空Biot数，在高维快速采样的同时提供后验精度保证。

2023-06-05cs.LGcs.NAmath.NA

XBART: Accelerated Bayesian Additive Regression Trees

University of Chicago（芝加哥大学）；Arizona State University（亚利桑那州立大学）

Bayesian additive regression trees (BART) (Chipman et. al., 2010) is a powerful predictive model that often outperforms alternative models at out-of-sample prediction. BART is especially well-suited to settings with unstructured predictor variables and substantial sources of unmeasured variation as is typical in the social, behavioral and health sciences. This paper develops a modified version of BART that is amenable to fast posterior estimation. We present a stochastic hill climbing algorithm that matches the remarkable predictive accuracy of previous BART implementations, but is many times …

2019-03-14cs.LGstat.ML

更多相关论文收起更多论文

Fast Posterior Estimation of Cardiac Electrophysiological Model Parameters via Bayesian Active Learning

Rochester Institute of Technology（罗切斯特理工学院）；Dalhousie University（达尔豪斯大学）；Johns Hopkins University（约翰斯·霍普金斯大学）

Probabilistic estimation of cardiac electrophysiological model parameters serves an important step towards model personalization and uncertain quantification. The expensive computation associated with these model simulations, however, makes direct Markov Chain Monte Carlo (MCMC) sampling of the posterior probability density function (pdf) of model parameters computationally intensive. Approximated posterior pdfs resulting from replacing the simulation model with a computationally efficient surrogate, on the other hand, have seen limited accuracy. In this paper, we present a Bayesian active lea…

2021-10-13cs.LG

Applying Physics-Informed Enhanced Super-Resolution Generative Adversarial Networks to Turbulent Non-Premixed Combustion on Non-Uniform Meshes and Demonstration of an Accelerated Simulation Workflow

Forschungszentrum Jülich GmbH（于利希研究中心）；RWTH Aachen University（亚琛工业大学）

This paper extends the methodology to use physics-informed enhanced super-resolution generative adversarial networks (PIESRGANs) for LES subfilter modeling in turbulent flows with finite-rate chemistry and shows a successful application to a non-premixed temporal jet case. This is an important topic considering the need for more efficient and carbon-neutral energy devices to fight the climate change. Multiple a priori and a posteriori results are presented and discussed. As part of this, the impact of the underlying mesh on the prediction quality is emphasized, and a multi-mesh approach is dev…

2022-10-28cs.LGphysics.flu-dyn

AcME -- Accelerated Model-agnostic Explanations: Fast Whitening of the Machine-Learning Black Box

Statwolf Data Science（Statwolf 数据科学公司）；Universitá degli Studi di Padova（帕多瓦大学）

In the context of human-in-the-loop Machine Learning applications, like Decision Support Systems, interpretability approaches should provide actionable insights without making the users wait. In this paper, we propose Accelerated Model-agnostic Explanations (AcME), an interpretability approach that quickly provides feature importance scores both at the global and the local level. AcME can be applied a posteriori to each regression or classification model. Not only does AcME compute feature ranking, but it also provides a what-if analysis tool to assess how changes in features values would affe…

2021-12-23cs.LG

在深度生成模型中分离内在模糊性与估计不确定性用于线性逆问题

Separating Intrinsic Ambiguity from Estimation Uncertainty in Deep Generative Models for Linear Inverse Problems

Neuroscience Institute, Carnegie Mellon University（卡内基梅隆大学神经科学研究所）；Machine Learning Department, Carnegie Mellon University（卡内基梅隆大学机器学习系）；Department of Electrical and Computer Engineering, Carnegie Mellon University（卡内基梅隆大学电气与计算机工程系）

本文提出一种分解后验不确定性的方法，用于区分逆问题中固有模糊性和推断传播的不确定性，通过校准分析揭示模型失败模式。

2026-05-14cs.LG

深度粗糙波动率中的不确定性与可解释性：一种神经信息论后验方法

Uncertainty and Explainability in Deep Rough Volatility: A Neural Information-Theoretic Posterior Approach

Imperial College（帝国理工学院）；King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）

本文提出一种基于模拟推理的粗糙Heston模型校准框架，利用神经比率估计学习参数后验分布，并引入Hellinger-SHAP可解释性方法，以提供不确定性感知的价格区间和参数信息增益分析。

2026-10-23cs.LGq-fin.PRstat.AP

加速的序列流匹配：从贝叶斯过滤视角

Accelerated Sequential Flow Matching: A Bayesian Filtering Perspective

Georgia Institute of Technology（佐治亚理工学院）

本文提出基于贝叶斯过滤的序列贝叶斯流匹配框架，通过学习概率流在时间步间传输后验分布，实现高效采样，提升实时流处理中的推断效率。

2026-05-13cs.LG

最大后验估计的可分级半定松弛

Scalable Semidefinite Relaxation for Maximum A Posterior Estimation

Stanford University（斯坦福大学）

离散马尔可夫随机场上的最大后验概率（MAP）推理是一个基本的任务，跨越了广泛的现实世界的应用，这是已知的一般图的NP-难。在本文中，我们提出了一种新的半定松弛公式（简称SDR）来估计MAP分配。在数学上，我们开发了一个加速的变体交替方向的乘法器（简称SDPAD-LR），可以有效地利用新的松弛的特殊结构。令人鼓舞的是，所提出的程序允许解决大规模问题的SDR，例如，网格图上的问题，包括成千上万的变量，每个节点有多个状态。与以前的SDP求解器相比，SDPAD-LR能够达到相当的精度，同时表现出显着提高的可扩展性，与通常认为的半定松弛只能应用于小规模的MRF问题。我们已经评估了SDR在各种基准数据集（包括OPENGM 2和PIC）上的性能，包括解决方案的质量和计算时间。实验结果表明，对于广泛的一类问题，SDPAD-LR优于国家的最先进的算法，以有效的方式产生更好的MAP分配。

2014-05-19cs.CVcs.ITcs.LG

Accelerated MM Algorithms for Ranking Scores Inference from Comparison Data

London School of Economics（伦敦政治经济学院）；KAIST（韩国科学技术院）

In this paper, we study a popular method for inference of the Bradley-Terry model parameters, namely the MM algorithm, for maximum likelihood estimation and maximum a posteriori probability estimation. This class of models includes the Bradley-Terry model of paired comparisons, the Rao-Kupper model of paired comparisons allowing for tie outcomes, the Luce choice model, and the Plackett-Luce ranking model. We establish tight characterizations of the convergence rate for the MM algorithm, and show that it is essentially equivalent to that of a gradient descent algorithm. For the maximum likeliho…

2020-12-26cs.LGmath.OCstat.ML

宇宙学基于似然推断的未来：加速高维参数估计与模型比较

The future of cosmological likelihood-based inference: accelerated high-dimensional parameter estimation and model comparison

Université de Genève（日内瓦大学）；University College London（伦敦大学学院）；Royal Holloway, University of London（伦敦大学皇家霍洛威学院）；Alan Turing Institute（阿兰·图灵研究所）

本文提出一种结合机器学习仿真模拟、可微分概率编程、基于梯度的MCMC采样和可扩展贝叶斯模型选择的宇宙学推断新范式，在高维参数空间中显著加速参数估计与模型比较，将计算时间从数月甚至数年缩短至数天。

2024-09-04astro-ph.COastro-ph.IMcs.LG

基于分布偏移检测的深度神经网络一次性在线测试

One-Shot Online Testing of Deep Neural Networks Based on Distribution Shift Detection

Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院（KIT））

针对忆阻交叉阵列上神经网络硬件加速器的测试需求，提出一种仅需单个测试向量的一次性在线测试方法，在大型拓扑和复杂任务中实现100%故障覆盖率，并显著降低内存开销和测试向量数量。

2023-05-16cs.AIcs.ETcs.LG