Here we propose a novel model family with the objective of learning to disentangle the factors of variation in data. Our approach is based on the spike-and-slab restricted Boltzmann machine which we generalize to include higher-order interactions among multiple latent variables. Seen from a generative perspective, the multiplicative interactions emulates the entangling of factors of variation. Inference in the model can be seen as disentangling these generative factors. Unlike previous attempts at disentangling latent factors, the proposed model is trained using no supervised information regarding the latent factors. We apply our model to the task of facial expression classification.
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。
Department of Applied Mathematics and Computer Science, Technical University of Denmark(丹麦技术大学应用数学与计算机科学系);Centre for Genomic Medicine, Rigshospitalet, Copenhagen University Hospital(哥本哈根大学医院Rigshospitalet基因组医学中心);Bioinformatics Centre, Department of Biology, University of Copenhagen(哥本哈根大学生物学系生物信息学中心)
Uncovering Physical Drivers of Dark Matter Halo Structures with Auxiliary-Variable-Guided Generative Models
Mathematics and Computer Science Division Argonne National Laboratory(阿贡国家实验室数学与计算机科学部);High Energy Physics Division Argonne National Laboratory(阿贡国家实验室高能物理部);Computational Science Division Argonne National Laboratory(阿贡国家实验室计算科学部);Computational Science Center Data, Analysis and Visualization Group National Renewable Energy Laboratory(国家可再生能源实验室计算科学中心数据、分析与可视化组)
Semi-Supervised Generative Models for Disease Trajectories: A Case Study on Systemic Sclerosis
University of Zurich(苏黎世大学);Oslo University Hospital(奥斯陆大学医院);University of Oslo(奥斯陆大学);University Hospital Zurich(苏黎世大学医院);Bogomolets National Medical University(博格莫列茨国立医科大学);CHU Lille(里尔大学医院中心);Lille University(里尔大学);University of Brescia(布雷西亚大学);Charles University(查理大学);University College London(伦敦大学学院);University of Split(斯普利特大学);University Hospital Center Split(斯普利特大学医院中心)
Modeling Complex Disease Trajectories using Deep Generative Models with Semi-Supervised Latent Processes
University of Zurich(苏黎世大学);ETH AI Center(苏黎世联邦理工学院人工智能中心);Oslo University Hospital(奥斯陆大学医院);University of Oslo(奥斯陆大学);University Hospital Zurich(苏黎世大学医院);Bogomolets National Medical University(博戈莫列茨国立医科大学)