期刊
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
卷 18, 期 -, 页码 2060-2075出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIFS.2023.3262112
关键词
Training; Privacy; Machine learning algorithms; Neural networks; Generative adversarial networks; Mathematical models; Loss measurement; Information-theoretic privacy; statistical inference; information bottleneck; obfuscation; generative models
Bottleneck problems are gaining attention in machine learning and information theory. In this work, the complexity-leakage-utility bottleneck (CLUB) model is proposed, which provides a unified theoretical framework, new interpretation of generative and discriminative models, insights for generative compression models, and fair generative models. The CLUB model is formulated as a complexity-constrained privacy-utility optimization problem and is connected to related problems such as information bottleneck (IB), privacy funnel (PF), and conditional entropy bottleneck (CEB). The deep variational CLUB (DVCLUB) models are introduced using neural networks and can be applied to generative models and fair representation learning problems. Extensive experiments are conducted on colored-MNIST and CelebA datasets.
Bottleneck problems are an important class of optimization problems that have recently gained increasing attention in the domain of machine learning and information theory. They are widely used in generative models, fair machine learning algorithms, design of privacy-assuring mechanisms, and appear as information-theoretic performance bounds in various multi-user communication problems. In this work, we propose a general family of optimization problems, termed as complexity-leakage-utility bottleneck (CLUB) model, which (i) provides a unified theoretical framework that generalizes most of the state-of-the-art literature for the information-theoretic privacy models, (ii) establishes a new interpretation of the popular generative and discriminative models, (iii) constructs new insights for the generative compression models, and (iv) can be used to obtain fair generative models. We first formulate the CLUB model as a complexity-constrained privacy-utility optimization problem. We then connect it with the closely related bottleneck problems, namely information bottleneck (IB), privacy funnel (PF), deterministic IB (DIB), conditional entropy bottleneck (CEB), and conditional PF (CPF). We show that the CLUB model generalizes all these problems as well as most other information-theoretic privacy models. Then, we construct the deep variational CLUB (DVCLUB) models by employing neural networks to parameterize variational approximations of the associated information quantities. Building upon these information quantities, we present unified objectives of the supervised and unsupervised DVCLUB models. Leveraging the DVCLUB model in an unsupervised setup, we then connect it with state-of-the-art generative models, such as variational auto-encoders (VAEs), generative adversarial networks (GANs), as well as the Wasserstein GAN (WGAN), Wasserstein auto-encoder (WAE), and adversarial auto-encoder (AAE) models through the optimal transport (OT) problem. We then show that the DVCLUB model can also be used in fair representation learning problems, where the goal is to mitigate the undesired bias during the training phase of a machine learning model. We conduct extensive quantitative experiments on colored-MNIST and CelebA datasets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据