针对生成模型相对参考模型的新颖性评估不足,本文提出谱差分聚类方法与 KEN 分数,量化多模态分布中的新模态,并在合成及真实图像数据上验证有效性。
摘要:
生成模型框架的大规模发展要求采用有原则的方法来评估模型相对于参考数据集的新颖性。尽管文献已广泛研究了生成模型的质量、多样性和泛化能力评估,但在机器学习领域,模型相对于参考模型的新颖性评估尚未得到充分探索。在本工作中,我们聚焦多模态分布的新颖性评估,并尝试解决以下差分聚类任务:给定生成模型 $P_\mathcal{G}$ 和参考模型 $P_\mathrm{ref}$ 的样本,我们如何发现 $P_\mathcal{G}$ 比 $P_\mathrm{ref}$ 更频繁表达的样本类型?我们为差分聚类任务引入一种谱方法,并提出基于核的熵新颖性(Kernel-based Entropic Novelty,KEN)分数,以量化 $P_\mathcal{G}$ 相对于 $P_\mathrm{ref}$ 的基于模态的新颖性。我们分析了具有良好可分成分的混合分布的 KEN 分数,并开发了一种基于核的方法来从经验数据计算 KEN 分数。我们通过在合成和真实图像数据集上的数值结果为 KEN 框架提供支持,结果表明该框架在检测新颖模态和比较生成模型方面的有效性。论文代码可在 www.github.com/buyeah1109/KEN 获取。
英文摘要:
The massive developments of generative model frameworks require principled methods for the evaluation of a model's novelty compared to a reference dataset. While the literature has extensively studied the evaluation of the quality, diversity, and generalizability of generative models, the assessment of a model's novelty compared to a reference model has not been adequately explored in the machine learning community. In this work, we focus on the novelty assessment for multi-modal distributions and attempt to address the following differential clustering task: Given samples of a generative model $P_\mathcal{G}$ and a reference model $P_\mathrm{ref}$, how can we discover the sample types expressed by $P_\mathcal{G}$ more frequently than in $P_\mathrm{ref}$? We introduce a spectral approach to the differential clustering task and propose the Kernel-based Entropic Novelty (KEN) score to quantify the mode-based novelty of $P_\mathcal{G}$ with respect to $P_\mathrm{ref}$. We analyze the KEN score for mixture distributions with well-separable components and develop a kernel-based method to compute the KEN score from empirical data. We support the KEN framework by presenting numerical results on synthetic and real image datasets, indicating the framework's effectiveness in detecting novel modes and comparing generative models. The paper's code is available at: www.github.com/buyeah1109/KEN
P-KDGAN: Progressive Knowledge Distillation with GANs for One-class Novelty Detection
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(中国科学院深圳先进技术研究院);School of Information and Electronics, Beijing Institute of Technology(北京理工大学信息与电子学院)
Plastic tensor networks for interpretable generative modeling
Institute for Solid State Physics, The University of Tokyo(东京大学固态物理研究所);Graduate School of Informatics, Kyoto University(京都大学信息科学研究生院);Institute for Physics of Intelligence, The University of Tokyo(东京大学智能物理研究所);Trans-scale Quantum Science Institute, The University of Tokyo(东京大学跨尺度量子科学研究所)
Generative and multimodal AI for materials prediction and design: Progress, challenges, and perspectives
School of Computer Science, University of Sheffield, Sheffield, UK(计算机科学学院,谢菲尔德大学);Centre for Machine Intelligence, University of Sheffield, Sheffield, UK(智能机器研究中心,谢菲尔德大学);School of Chemical, Materials and Biological Engineering, University of Sheffield, Sheffield, UK(化学、材料与生物工程学院,谢菲尔德大学);Henry Royce Institute, Royce Discovery Centre, University of Sheffield, Sheffield, UK(亨利·罗伊奇研究所,谢菲尔德大学);Materials Nexus Ltd., Salisbury House, Cambridge, UK(材料 nexus 有限公司,剑桥,英国);School of Mathematical and Physical Sciences, University of Sheffield, Sheffield, UK(数学与物理科学学院,谢菲尔德大学)
Disentangling Interpretable Generative Parameters of Random and Real-World Graphs
University College London(伦敦大学学院);Microsoft Research Cambridge(微软剑桥研究院)
虽然存在各种各样的可解释的图形生成程序,匹配观察到的图形拓扑结构与这些程序和选择其参数仍然是一个悬而未决的问题。设计生成模型,密切再现现实世界的图形需要领域知识和耗时的模拟。虽然现有的深度学习方法依赖于较少的手动建模,但它们几乎没有提供可解释性。这项工作的方法图生成(解码)的逆图压缩(编码)。我们表明,在一个以解纠缠为重点的深度自动编码框架中,特别是Beta-变分自编码器(Beta-VAE),生成过程及其参数的选择自然出现在潜在空间中。我们的模型能够学习解纠缠的,可解释的潜变量,这些潜变量代表程序生成的随机图和真实世界的图的生成参数。解纠缠的程度是使用互信息间隙(Mutual Information Gap,简称ISI)来定量测量的。当在ER随机图上训练我们的Beta-VAE模型时,它的潜变量与ER随机图参数n和p有一个接近一对一的映射。我们部署该模型来分析图拓扑和节点属性之间的相关性,测量它们的相互依赖性,而无需手工挑选拓扑属性。
Interpretable Generative and Discriminative Learning for Multimodal and Incomplete Clinical Data
Department of Health Sciences and Technology (DHEST), ETH Zurich(健康科学与技术系,苏黎世联邦理工学院);Department of Signal Theory and Communications, Universidad Carlos III de Madrid(信号理论与通信系,卡洛斯三世大学);UCL Hawkes Institute, Department of Computer Science, University College London(UCL Hawkes研究所,伦敦大学学院计算机科学系);Instituto de Investigación Sanitaria Gregorio Marañón (IiSGM), Madrid(Gregorio Marañón卫生研究所(IiSGM),马德里)
LVNS-RAVE: Diversified audio generation with RAVE and Latent Vector Novelty Search
University of Oslo(奥斯陆大学);Institute for Clarity in Documentation(文献清晰度研究所);Inria Paris-Rocquencourt(法国国家信息与自动化研究所巴黎-罗康分校);Rajiv Gandhi University(拉吉夫·甘地大学);Tsinghua University(清华大学);Palmer Research Laboratories(帕尔默研究实验室)
Restyling Unsupervised Concept Based Interpretable Networks with Generative Models
Sorbonne Université(索邦大学);Technical University of Munich(慕尼黑工业大学);Helmholtz Munich(亥姆霍兹慕尼黑中心);Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心);LTCI, Télécom Paris, Institut Polytechnique de Paris(巴黎理工学院 巴黎高等电信学院 LTCI)
A Lightweight Generative Model for Interpretable Subject-level Prediction
Technical University of Denmark(丹麦技术大学);Massachusetts General Hospital(麻省总医院);Harvard Medical School(哈佛医学院);Copenhagen University Hospital Hvidovre(希维多夫哥本哈根大学医院);Technical University of Munich(慕尼黑工业大学);Aalto University(阿尔托大学)
Weighted Support Points from Random Measures: An Interpretable Alternative for Generative Modeling
Department of Mathematics and Statistics and Data Sciences, University of Texas at Austin(数学与统计学及数据科学系,德克萨斯大学奥斯汀分校);Department of Probability and Statistics, IIMAS, Universidad Nacional Autónoma de México(概率与统计系,墨西哥国家自治大学)
Interpretable epistemic uncertainty decomposition in sequential generative models via polynomial chaos surrogates
Mathematical Institute, University of Oxford, United Kingdom(牛津大学数学研究所,英国);IBM Research, Thomas J. Watson Research Center, USA(IBM研究院,托马斯·J·沃森研究中心,美国);IBM Research Europe, Daresbury, United Kingdom(IBM欧洲研究院,达尔斯伯里,英国);MIT–IBM Watson AI Lab, Massachusetts, USA(麻省理工–IBM沃森人工智能实验室,马萨诸塞州,美国);Université de Montréal, Mila–Quebec AI Institute, Montreal, Quebec, Canada(蒙特利尔大学,魁北克人工智能研究所,蒙特利尔,魁北克,加拿大)
Decontextualized learning for interpretable hierarchical representations of visual patterns
Max Planck Institute of Animal Behavior(马克斯·普朗克动物行为研究所);University of Konstanz(康斯坦茨大学);Centro de Investigaciones Científicas de las Huastecas Aguazarca, A.C.(华斯特卡斯阿瓜扎尔卡科学研究中心);University of Würzburg(维尔茨堡大学);Texas A&M University(得克萨斯农工大学);Texas State University San Marcos(得克萨斯州立大学圣马科斯分校)
Fractal Flow: Hierarchical and Interpretable Normalizing Flow via Topic Modeling and Recursive Strategy
School of Mathematics Harbin Institute of Technology(哈尔滨工业大学数学学院);School of Mathematics and Institute for Artificial Intelligence Harbin Instiute of Technology(哈尔滨工业大学数学学院和人工智能研究院);School of Earth and Space Sciences and Institute for Artificial Intelligence Peking University(北京大学地球和空间科学学院和人工智能研究院)
Interpretable Reward Redistribution in Reinforcement Learning: A Causal Approach
Eindhoven University of Technology(埃因霍温理工大学);University of California San Diego(加州大学圣迭戈分校);University College London(伦敦大学学院);University of Liverpool(利物浦大学)
VAE-Inf: A statistically interpretable generative paradigm for imbalanced classification
Department of Data Science and Artificial Intelligence(数据科学与人工智能系);The Hong Kong Polytechnic University(香港理工大学);Department of Applied Mathematics(应用数学系)
Interpretable Imitation Learning via Generative Adversarial STL Inference and Control
Department of Mechanical Engineering(机械工程系);Boston University(波士顿大学);Computer Science and Artificial Intelligence Laboratory(计算机科学与人工智能实验室);MIT(麻省理工学院);Department of Electrical and Computer Engineering(电气与计算机工程系);University of Maryland(马里兰大学)