arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.10623cs.LGstat.ML
收藏

从相关数据中学习结构化潜在因子:一个信息论视角的生成模型框架

Learning Structured Latent Factors from Dependent Data:A Generative Model Framework from Information-Theoretic Perspective

Mila(米拉);Université de Montréal(蒙特利尔大学);Preferred Networks Inc.(优选网络公司)

Ruixiang Zhang, Masanori Koyama, Katsuhiko Ishiguro

摘要:

学习具有所需结构属性的多变量数据的可控和可推广的表示仍然是机器学习中的基本问题。在本文中,我们提出了一个新的框架,学习生成模型与各种潜在的空间结构。我们以掩码变量的形式表示归纳偏差,以在图模型中对依赖结构进行建模,并扩展了多元信息瓶颈理论来执行它。我们的模型提供了一种原则性的方法来学习一组语义上有意义的潜在因素,这些因素反映了各种类型的期望结构,如捕获相关性或编码不变性,同时还提供了根据数据自动估计依赖性结构的灵活性。我们表明,我们的框架统一了许多现有的生成模型,可以应用于各种任务,包括多模态数据建模,算法公平性,不变的风险最小化。

英文摘要:

Learning controllable and generalizable representation of multivariate data with desired structural properties remains a fundamental problem in machine learning. In this paper, we present a novel framework for learning generative models with various underlying structures in the latent space. We represent the inductive bias in the form of mask variables to model the dependency structure in the graphical model and extend the theory of multivariate information bottleneck to enforce it. Our model provides a principled approach to learn a set of semantically meaningful latent factors that reflect various types of desired structures like capturing correlation or encoding invariance, while also offering the flexibility to automatically estimate the dependency structure from data. We show that our framework unifies many existing generative models and can be applied to a variety of tasks including multi-modal data modeling, algorithmic fairness, and invariant risk minimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑