Bayesian Inverse Problems Meet Flow Matching: Efficient and Flexible Inference via Transformers
Skolkovo Institute of Science and Technology(斯克洛夫学院科学与技术研究所);Sberbank(储蓄银行);AI4S Center(AI4S中心);Artificial Intelligence Research Institute(人工智能研究 institute)
Daniil Sherki, Ivan Oseledets, Ekaterina Muravleva
The efficient resolution of Bayesian inverse problems remains challenging due to the high computational cost of traditional sampling methods. In this paper, we propose a novel framework that integrates Conditional Flow Matching (CFM) with a transformer-based architecture to enable fast and flexible sampling from complex posterior distributions. The proposed methodology involves the direct learning of conditional probability trajectories from the data, leveraging CFM's ability to bypass iterative simulation and transformers' capacity to process arbitrary numbers of observations. The efficacy of the proposed framework is demonstrated through its application to three problems: a simple nonlinear model, a disease dynamics framework, and a two-dimensional Darcy flow Partial Differential Equation. The primary outcomes demonstrate that the relative errors in parameters recovery are as low as 1.5%, and that the inference time is reduced by up to 2000 times on CPU in comparison with the Monte Carlo Markov Chain. This framework facilitates the expeditious resolution of Bayesian problems through the utilisation of sampling from the learned conditional distribution.
Gaussian processes meet NeuralODEs: A Bayesian framework for learning the dynamics of partially observed systems from scarce and noisy data
University of Pennsylvania(宾夕法尼亚大学)
本文提出了一个机器学习框架(GP-NODE),用于从非线性动态系统的部分、噪声和不规则观测中识别贝叶斯系统。该方法利用微分编程的最新发展,通过常微分方程求解器传播梯度信息,并使用Hamilton Monte Carlo采样和高斯过程先验对所观察到的系统状态进行贝叶斯推断。这使我们能够利用观测数据中的时间相关性,并有效地推断具有量化不确定性的合理模型的后验分布。此外,使用稀疏促进先验,如芬兰马蹄免费模型参数,使潜在的潜在动态的可解释和简约的表示发现。一系列的数值研究表明,建议的GP-NODE方法的有效性,包括捕食系统,系统生物学,和50维人体运动动力系统。总之,我们的研究结果提出了一种新的,灵活的和强大的工作流程,数据驱动的模型发现下的不确定性。本手稿随附的所有代码和数据可在\url{https://github.com/PredictiveIntelligenceLab/GP-NODEs}在线获取。
When Bayesian Tensor Completion Meets Multioutput Gaussian Processes: Functional Universality and Rank Learning
College of Information Science and Electronic Engineering, Zhejiang University(浙江大学信息科学与电子工程学院);Zhejiang Provincial Key Laboratory of Multi-Modal Communication Networks and Intelligent Information Processing(浙江省多模态通信网络与智能信息处理重点实验室);School of Science & Engineering, The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)科学与工程学院);Department of Electrical and Electronic Engineering, The University of Hong Kong(香港大学电子与电气工程系);Technical University of Denmark(丹麦技术大学);NoiseLab, University of California, San Diego(加州大学圣地亚哥分校NoiseLab);HERON - Center of Excellence in Robotics, Athena R.C(HERON-机器人卓越中心)
Real-Time Optimization Meets Bayesian Optimization and Derivative-Free Optimization: A Tale of Modifier Adaptation
Imperial College London(帝国理工学院);University College London(伦敦大学学院);Norwegian University of Science and Technology(挪威科技大学);Radix Engenharia e Software(Radix工程与软件公司);Universidade de São Paulo(圣保罗大学)
A flexible Bayesian non-parametric mixture model reveals multiple dependencies of swap errors in visual working memory
Department of Engineering University of Cambridge(工程系剑桥大学);Department of Psychology University of Cambridge(心理学系剑桥大学);Department of Cognitive Science Central European University(认知科学系中央欧亚大学)