正态均值估计中收缩阈值规则的近似风险最小化研究

近日，arXiv预印本收录了一篇题为“Approximate Risk Minimization Over Shrinking-Thresholding Rules in Normal Mean Estimation”的论文，作者为Wei Jiang。该研究聚焦于正态均值估计这一统计学基础问题，探讨在收缩阈值规则（如软阈值、硬阈值等）下如何实现近似风险最小化。收缩阈值方法在高维数据分析和信号处理中应用广泛，论文可能提出了新的理论框架或风险准则，以优化此类规则的选择和性能。预印本于2026年7月7日提交，为相关领域提供了新的理论视角。 #论文 #arXiv #统计学 #收缩阈值 #正态均值估计 #风险最小化 #高维统计 #预印本

Mengwu Guo 提出微分方程高斯过程逼近的统一视角

研究人员 Mengwu Guo 于2026年7月7日在arXiv预印本平台提交了一篇题为“A unified perspective of Gaussian process approximation for differential equations”的论文。该论文旨在为高斯过程在微分方程求解中的应用提供统一的理论视角，系统梳理并连接了现有多种逼近方法。高斯过程作为一种非参数贝叶斯工具，在科学计算与机器学习领域日益受到关注，而这篇论文提出的统一框架有助于加深对不同方法的理解，并可能推动新的算法设计。目前论文已发布PDF全文和HTML预览，供研究者自由获取。 #高斯过程 #微分方程 #统一视角 #arXiv #科学计算 #机器学习 #数值方法 #论文 #计算数学

张量程序定量高斯过程极限论文在arXiv发布

Andrea Agazzi等三位作者在arXiv预印本平台提交了题为《Quantitative Gaussian-Process limits of Tensor Programs》的论文。该论文于2026年7月7日发布，聚焦于张量程序的高斯过程极限的定量分析，属于机器学习理论前沿研究，涉及计算机科学、数学与统计学等多个学科。论文目前以PDF形式开放获取，为理解深度神经网络的初始化行为与泛化理论提供了新的定量视角。 #arXiv #张量程序 #高斯过程 #机器学习 #深度学习 #预印本 #科技新闻

Closed-form fractional radial links for elliptical Mahalanobis discriminant analysis

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Serhii Zabolotnii Dr. [ view email ] [v1] Tue, 7 Jul 2026 10:02:33 UTC (48 KB) Full-text links: Access Paper: View a PDF of the paper titled Closed-form fractional radial links for elliptical Mahalanobis discriminant analysis, by Serhii Zabolotnii View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: math stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connecte

d Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

研究提出基于冰碛物数据的冰川前缘轨迹随机生成方法

一项发表于arXiv预印本的研究提出了一种基于记录数据的轨迹随机生成器，并将其应用于冰川前缘位置波动的模拟分析。冰川前缘的进退是气候变化的重要指示器，但长期连续观测数据较为稀缺。研究人员利用冰碛物样本中保存的历史记录，通过随机生成模型重建冰川位置的潜在变化轨迹，从而量化波动的不确定性并揭示其动态模式。该工作由法国气候与环境科学实验室（LSCE）的Maud MEGRET等人完成，为古气候代用资料的应用和冰川动力学研究提供了新思路。 #冰川 #随机生成器 #冰碛物 #气候模型 #轨迹 #论文 #预印本 #古气候 #地球科学

没有子空间可追踪

据arXiv预印本平台显示，一篇题为“No Subspace to Track: Non-Identifiability and Optimizer State in Low-Rank Training”的论文于2026年7月7日提交。该论文由Noel Thomas撰写，直接针对低秩训练方法中的一个核心理论难题——非可识别性（non-identifiability）进行探讨，并分析了优化器状态（optimizer state）在这一问题中的关键作用。低秩训练是当前高效模型压缩与加速的重要手段，但该研究指出，传统观点下试图追踪特定子空间的做法可能无法实现，这为理解低秩训练的收敛行为与模型性能提供了新的理论视角。论文的发现对于未来优化低秩训练算法、改进参数更新策略具有重要的指导意义。 #低秩训练 #非可识别性 #优化器状态 #机器学习 #深度学习 #arXiv论文 #模型压缩 #理论分析

层次贝叶斯模型混合权重后验分布精确计算论文发布

日前，研究者Georgy Meshcheryakov在arXiv预印本平台提交了一篇题为“Exact computation of posterior distribution of mixture weights in hierarchical Bayesian models”的论文。该论文围绕层次贝叶斯模型中混合权重的后验分布计算问题，提出了一种精确求解的方法。层次贝叶斯模型在机器学习与统计推断中应用广泛，混合权重的估计是其中的核心环节。传统方法多依赖于近似采样，而该论文的方案

组合学习的函数空间二分法

Arkaprabha Ganguli等人在arXiv发表最新论文，提出组合学习的函数空间二分法，从理论上严格证明神经正切核在需要组合多个子任务的学习场景中，相比最优解存在指数级别的次优性。该研究揭示了NTK在处理复杂结构化问题时的固有局限，为理解深度学习表示效率与理论边界提供了新视角。论文于2026年7月提交。 #神经正切核 #组合学习 #深度学习 #理论 #机器学习 #AI #研究 #arXiv

Telegram必备的搜索引擎，极搜JISOU帮你精准找到，想要的群组、频道、视频、音乐

👉 t.me/jisou2?start=a_8247614025

凸近似框架论文

近日，Fabian Schneider、Tapio Helin 和 Leila Taghizadeh 在 arXiv 预印本平台提交了一篇题为《A Convex Approximation Framework for Neural Likelihood-Based Bayesian Inverse Problems》的论文。该研究提出了一种新颖的凸近似框架，专门用于处理基于神经似然的贝叶斯逆问题。贝叶斯逆问题在科学计算、工程和机器学习等领域扮演关键角色，但求解过程常涉及高维复杂分布，计算成本高昂。作者通过引入凸优化技术，在保持近似精度的同时显著提升了求解的稳健性和效率。论文还讨论了该框架与现有神经似然模型（如归一化流等）的兼容性，为不确定量化和高维参数推断提供了新的理论工具。目前该论文已开放预览，相关代码和数据可能后续发布，以推动该领域的发展。 #论文 #贝叶斯逆问题 #凸近似 #神经网络 #机器学习 #arXiv #科学计算 #不确定性量化 #优化算法

Separation Capacity of Scattering Networks on Low-Dimensional Datasets

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Konstantin Häberle [ view email ] [v1] Tue, 7 Jul 2026 09:25:58 UTC (23 KB) Full-text links: Access Paper: View a PDF of the paper titled Separation Capacity of Scattering Networks on Low-Dimensional Datasets, by Konstantin H\"aberle and Helmut B\"olcskei View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs math stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) C

onnected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

对称散度下图拉普拉斯收敛性研究在arXiv发布

近日，一篇题为《On the convergence of graph Laplacians with a symmetric divergence》的学术论文在预印本平台arXiv上公开。该论文由Liane Xu独立撰写，于2026年7月7日提交。图拉普拉斯算子是图论与机器学习领域的重要数学工具，广泛应用于谱聚类、图信号处理及图神经网络等方向。本研究探讨在对称散度条件下图拉普拉斯算子的收敛特性，为理解图结构数据的深层几何提供理论支撑。目前论文全文以PDF格式开放，可供相关研究者下载参考，预计将对流形学习与图深度学习等前沿领域产生推动作用。 #学术论文 #arXiv #图拉普拉斯 #对称散度 #收敛性 #机器学习 #图深度学习 #LianeXu

Boosting with List

Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Addison Prairie [ view email ] [v1] Tue, 7 Jul 2026 03:34:53 UTC (36 KB) Full-text links: Access Paper: View a PDF of the paper titled Boosting with List-Decodable Codes, by Addison Prairie and 1 other authors View PDF HTML (experimental) TeX Source view license Current browse context: < prev | next > new | recent | 2026-07 Change to browse by: cs stat References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? )

Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle ( What is ? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

平均场贝叶斯神经网络宽度鲁棒可学习性研究公开

近日，一篇题为《平均场贝叶斯神经网络中的宽度鲁棒可学习性》的学术论文在预印本平台arXiv上提交。该论文由Dmitry Vaintrob和Kaarel Hanni共同撰写，于2026年7月7日发布。研究聚焦于平均场贝叶斯神经网络中网络宽度对模型可学习性的鲁棒性影响，通过理论分析探讨了宽度参数如何影响模型的泛化能力与训练稳定性。该工作旨在深化对贝叶斯神经网络理论的理解，为设计更可靠的深度概率模型提供新视角。目前论文以预印本形式公开，提供PDF和HTML版本全文访问。 #arXiv #论文 #贝叶斯神经网络 #平均场 #宽度鲁棒 #可学习性 #机器学习 #深度学习 #AI #理论研究

亚马逊云科技发布《企业生产级智能体开发部署指南》 以评估驱动Agent部署新范式

亚马逊云科技在2026中国峰会上发布《企业生产级智能体开发部署指南》。当前Agentic AI迎来爆发拐点，但企业将智能体从原型推向生产面临巨大挑战，传统评估方法因大模型概率性输出而失效，导致大量项目停滞或取消。指南提出以评估驱动的开发方法论，涵盖定标准、开发、评估、灰度上线等六个步骤，并覆盖八类评估维度及自动化评估框架，帮助企业构建从底层模型到业务结果的可观测、可持续改进闭环。该指南融合亚马逊内部实践，已在购物助手、客服、卖家等场景深度应用，并助力多家行业客户突破规模化部署瓶颈，实现真实业务价值。 #亚马逊云科技 #AgenticAI #智能体 #评估驱动 #企业部署 #AI #云计算