An important subclass of hybrid Bayesian networks are those that represent Conditional Linear Gaussian (CLG) distributions --- a distribution with a multivariate Gaussian component for each instantiation of the discrete variables. In this paper we explore the problem of inference in CLGs. We show that inference in CLGs can be significantly harder than inference in Bayes Nets. In particular, we prove that even if the CLG is restricted to an extremely simple structure of a polytree in which every continuous node has at most one discrete ancestor, the inference task is NP-hard.To deal with the often prohibitive computational cost of the exact inference algorithm for CLGs, we explore several approximate inference algorithms. These algorithms try to find a small subset of Gaussians which are a good approximation to the full mixture distribution. We consider two Monte Carlo approaches and a novel approach that enumerates mixture components in order of prior probability. We compare these methods on a variety of problems and show that our novel algorithm is very promising for large, hybrid diagnosis problems.
MIT CSAIL(麻省理工学院计算机科学与人工智能实验室);Binghamton University(宾汉姆顿大学);Université Paris Cité(巴黎城市大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);IRIF(法国信息学研究所);IBM Research(IBM研究院)
Research Focus Cognitive Sciences(认知科学研究重点中心);University of Potsdam(波茨坦大学);Artificial Intelligence Lab(人工智能实验室);Otto von Guericke University(奥托·冯·格里克大学)
An Incremental Explanation of Inference in Hybrid Bayesian Networks for Increasing Model Trustworthiness and Supporting Clinical Decision Making
School of Electronic Engineering & Computer Science, Queen Mary University of London(伦敦玛丽女王大学电子工程与计算机科学学院);Centre for Trauma Sciences, Blizard Institute, Queen Mary University of London(伦敦玛丽女王大学布利扎德研究所创伤科学中心);Royal London Hospital, Barts and The London NHS Trust(皇家伦敦医院);Academic Department of Military Surgery and Trauma, Royal Centre for Defence Medicine(皇家国防医学中心军事外科与创伤学术部)
Hybrid Probabilistic Inference with Logical Constraints: Tractability and Message Passing
University of California, LA(加利福尼亚大学洛杉矶分校);AMSS(数学与系统科学研究院);Chinese Academy of Sciences(中国科学院);DISI(信息工程与计算机科学系);University of Trento(特伦托大学)
加权模型集成(Weighted Model Integration,简称WMM)是一个非常吸引人的概率推理框架:它允许通过可满足性模理论(Satisfiability Modulo Theories,简称SMT)的语言来表达真实世界混合场景的复杂依赖关系,其中变量在本质上是异构的(包括连续和离散);以及计算具有任意复杂逻辑约束的概率查询。最近的工作表明,在某些假设下,可简化的模型集成(MI)问题,从而有效地允许混合概率推理的体积计算。在本文中,我们介绍了一种新的制定MI通过消息传递计划,允许有效地计算的边际密度和统计矩的所有变量在线性时间。因此,当任意丰富的MI查询符合问题结构时,我们能够摊销推理,这里表示为与SMT公式相关联的原始图。此外,我们从理论上跟踪的易处理边界的确切MI。事实上,我们证明,在原始图的结构要求,使我们的MI算法易于处理-边界的直径和树宽-边界不仅是足够的,但通过MI易于推理的必要条件。
HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference
The State Key Laboratory of Blockchain and Data Security, Zhejiang University(浙江大学区块链与数据安全国家重点实验室);Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security(杭州高新区(滨江)区块链与数据安全研究院);Hangzhou Dianzi University(杭州电子科技大学)
Scaling up Hybrid Probabilistic Inference with Logical and Arithmetic Constraints via Message Passing
University of California, Los Angeles(加州大学洛杉矶分校);University of Trento(特伦托大学);Academy of Mathematics and Systems Science, Chinese Academy of Sciences(中国科学院数学与系统科学研究院)
加权模型集成(Weighted Model Integration,简称WIM)是一个非常吸引人的概率推理框架:它允许通过可满足性模理论(Satisfiability Modulo Theories,简称SMT)的语言来表达变量既连续又离散的真实世界问题的复杂依赖关系,以及计算具有复杂逻辑和算术约束的概率查询。然而,现有的非线性规划求解器还没有准备好扩展到这些问题。它们要么完全忽略了问题的内在依赖结构,要么局限于限制性太强的结构。为了缩小这一差距,我们推导出一个因式分解的形式主义,使我们能够设计一个可扩展的基于消息传递,MP-BSSOVER。也就是说,MP-SVM是第一个SVM求解器,它允许:1)对整个树结构的SVM问题进行精确推理; 2)在线性时间内计算所有边缘密度; 3)摊销查询间的推理。实验结果表明,我们的求解器显着优于现有的非线性规划求解器在一个大的基准集。
HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning
East China Normal University(华东师范大学);Beihang University(北京航空航天大学);Shanghai University of International Business and Economics(上海对外经贸大学);Fudan University(复旦大学)
HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling
Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology(麻省理工学院计算机科学与人工智能实验室);Toyota Research Institute(丰田研究院);Massachusetts Institute of Technology(麻省理工学院);Computer Science and Artificial Intelligence Laboratory(计算机科学与人工智能实验室)
$χ$SPN: Characteristic Interventional Sum-Product Networks for Causal Inference in Hybrid Domains
Indian Institute of Technology Bombay(印度孟买理工学院);Technical University of Darmstadt(达姆施塔特工业大学);Hessian Center for Artificial Intelligence (hessian.AI)(黑森人工智能中心);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心);Eindhoven University of Technology(埃因霍温理工大学)
Towards Practical Multi-Robot Hybrid Tasks Allocation for Autonomous Cleaning
Harbin Institute of Technology(哈尔滨工业大学);Shenzhen Institute of Advanced Technology, Chinese Academy of Science(中国科学院深圳先进技术研究院);Chongqing University(重庆大学);Gaussian Robotics(高仙机器人)
Practical hybrid PQC-QKD protocols with enhanced security and performance
Pritzker School of Molecular Engineering, The University of Chicago(芝加哥大学普利兹克分子工程学院);Materials Science Division, Argonne National Laboratory(阿贡国家实验室材料科学部);Center for Molecular Engineering, Argonne National Laboratory(阿贡国家实验室分子工程中心);Department of Computer Science, The University of Chicago(芝加哥大学计算机科学系);SeQure;College of Information and Computer Sciences, University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校信息与计算机科学学院)