Diffusion models generate high-quality synthetic data. They operate by defining a continuous-time forward process which gradually adds Gaussian noise to data until fully corrupted. The corresponding reverse process progressively "denoises" a Gaussian sample into a sample from the data distribution. However, generating high-quality outputs requires many discretization steps to obtain a faithful approximation of the reverse process. This is expensive and has motivated the development of many acceleration methods. We propose to accomplish sample generation by learning the posterior {\em distribution} of clean data samples given their noisy versions, instead of only the mean of this distribution. This allows us to sample from the probability transitions of the reverse process on a coarse time scale, significantly accelerating inference with minimal degradation of the quality of the output. This is accomplished by replacing the standard regression loss used to estimate conditional means with a scoring rule. We validate our method on image and robot trajectory generation, where we consistently outperform standard diffusion models at few discretization steps.
Generative Adversarial Nets for Robust Scatter Estimation: A Proper Scoring Rule Perspective
University of Chicago（芝加哥大学）；Hong Kong University of Science and Technology（香港科技大学）
稳健散布估计是统计学中的一项基本任务。Gao et al.（2018）最近关于鲁棒估计和生成对抗网络（GANs）之间的联系的发现表明，可以使用优化GANs的类似技术来计算类似深度的鲁棒估计器。在本文中，我们介绍了一个一般的学习通过分类框架的基础上适当的评分规则的概念。这个框架允许我们通过$f$的变分近似的透镜来理解矩阵深度函数和各种GANs-由适当的评分规则引起的发散。然后，我们提出了一类新的强大的散布估计在这个框架中仔细构造鉴别器与适当的神经网络结构。在Huber污染模型下，证明了这些估计量能达到极小极大散布率估计.我们的数值结果表明，在各种设置下对文献中的竞争对手的良好性能。
School of Computer Science, University of Birmingham, Birmingham, UK GSK Drug Metabolism \& Pharmacokinetics, GSK Medicines Research Centre, Stevenage, UK GSK Genome Biology, Collegeville, PA, USA School of Electronics, Electrical Engineering；Computer Science, Queen's University Belfast, UK
Probabilistic Forecasting with Generative Networks via Scoring Rule Minimization
University of Oxford（牛津大学）；Southern University of Science and Technology（南方科技大学）；University of Warwick（华威大学）；European Centre for Medium-Range Weather Forecasts（欧洲中期天气预报中心）
Learning Optimal Distributionally Robust Individualized Treatment Rules Integrating Multi-Source Data
Department of Applied Mathematics, The Hong Kong Polytechnic University（应用数学系，香港理工大学）；Department of Biostatistics, City University of Hong Kong（生物统计学系，香港城市大学）
Uncertainty Quantification for Regression using Proper Scoring Rules
Department of Machine Learning, MBZUAI（机器学习部门，MBZUAI）；ELLIS Unit Linz and LIT AI Lab, Institute for Machine Learning, Johannes Kepler University Linz（林兹ELLIS单元和LIT AI实验室，机器学习研究所，林兹约翰内斯·开普勒大学）
Likelihood-Free Inference with Generative Neural Networks via Scoring Rule Minimization
University of Oxford（牛津大学）；University of Warwick（华威大学）
贝叶斯似然自由推理方法产生后验近似的模拟器模型与棘手的可能性。近年来，许多工作训练神经网络来逼近难以处理的似然或后验。大多数建议使用归一化流，即神经网络参数化可逆映射，用于从底层基本度量转换样本;然后可以访问转换样本的概率密度，并且可以通过模拟参数-观测对的极大似然训练归一化流。最近的一项工作[拉梅什等人，2022]用生成网络近似后验，这降低了可逆性要求，因此是一类更灵活的分布，可以扩展到高维和结构化数据。然而，生成网络只允许从参数化分布中采样;因此，Ramesh et al.[2022]遵循对抗训练的常见解决方案，其中生成网络与"批评者"网络进行最小-最大游戏。这个过程是不稳定的，可能导致一个学习的分布低估了不确定性-在极端情况下崩溃到一个点。在这里，我们建议使用通过评分规则最小化训练的生成网络来近似后验，这是一种被忽视的无对抗方法，可以实现平滑训练和更好的不确定性量化。在模拟研究中，评分规则方法在对抗框架方面具有更短的训练时间，从而产生更好的性能。
Logical Expressivity and Explanations for Monotonic GNNs with Scoring Functions
Department of Compute Science, University of Oxford（牛津大学计算机科学系）；Department of Computer Science, Royal Holloway, University of London（伦敦大学皇家霍洛威学院计算机科学系）
Proper Scoring Rules for Right-Censored Survival Data
IDLab；Biometrics Research Group（生物信息学研究组）；Department of Electronics（电子系）；Department of Morphology, and Information Systems and Information Systems Imaging, Orthopedics（形态学系及信息系统和影像学、骨科）；Ghent University（根特大学）；Ghent University - imec（根特大学 - imec）；Rehabilitation and Nutrition（康复与营养）；Department of Electronics and Information Systems（电子系及信息系统）
STIPP: Space-time in situ postprocessing over the French Alps using proper scoring rules
Inria, Paris, France（法国巴黎的Inria）；Météo-France, CNRS, Univ. Grenoble Alpes, Univ. Toulouse, CNRM, Centre d’Études de la Neige, 38000 Grenoble, France（法国气象局、国家科学研究中心、格勒诺布尔阿尔卑斯大学、图卢兹大学、CNRM、雪研究中心，法国格勒诺布尔）；University of Colorado Boulder, Boulder, United States（科罗拉多大学博尔德分校）
Beyond Discriminant Patterns: On the Robustness of Decision Rule Ensembles
School of Computing；Information Technology Great Bay University Dongguan, China；School of Informatics University of Edinburgh Edinburgh, UK；DAI Cluster Eindhoven University of Technology Eindhoven, Netherlands；Department of Computer Science Mohamed bin Zayed University of Artificial Intelligence Abu Dhabi, UAE；Department of Mathematics；Computer Science Eindhoven University of Technology Eindhoven, Netherlands
