A Communication Efficient Collaborative Learning Framework for Distributed Features

Liu, Yang; Kang, Yan; Zhang, Xinwei; Li, Liping; Cheng, Yong; Chen, Tianjian; Mei, Hong; Yang, Qiang

doi:10.48550/arxiv.1912.11187

Cited by 25 publications

(44 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The other algorithm is the Federated Stochastic Block Coordinate Descent(FedBCD) algorithm that has a similar principle with FedAvg, targets total rounds of communications, skipping updates for each iteration to achieve desired accuracy rate [25].…”

Section: Centralized Aggregationmentioning

confidence: 99%

Secure Multi-Party Computation based Privacy Preserving Data Analysis in Healthcare IoT Systems

ŞAHİNBAŞ¹,

Çatak²

2021

Preprint

View full text Add to dashboard Cite

Recently, many innovations have been experienced in healthcare by rapidly growing Internet-of-Things (IoT) technology that provides significant developments and facilities in the health sector and improves daily human life. The IoT bridges people, information technology and speed up shopping. For these reasons, IoT technology has started to be used on a large scale. Thanks to the use of IoT technology in health services, chronic disease monitoring, health monitoring, rapid intervention, early diagnosis and treatment, etc. facilitates the delivery of health services. However, the data transferred to the digital environment pose a threat of privacy leakage. Unauthorized persons have used them, and there have been malicious attacks on the health and privacy of individuals. In this study, it is aimed to propose a model to handle the privacy problems based on federated learning. Besides, we apply secure multi party computation. Our proposed model presents an extensive privacy and data analysis and achieve high performance.

show abstract

Section: Centralized Aggregationmentioning

confidence: 99%

Secure Multi-Party Computation based Privacy Preserving Data Analysis in Healthcare IoT Systems

ŞAHİNBAŞ¹,

Çatak²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…, X K ) and the scalar output label Y . As illustrated in Figure 1, each kth feature X k in vector X is observed only at the kth agent, while the output label Y is observed at all the K agents [5], [6]. Features and labels can take values in arbitrary alphabets.…”

Section: Problem Formulationmentioning

confidence: 99%

“…is sufficiently large to include the posterior distribution P Y | X ,X k ,Y, X,X k . In fact, for any fixed aggregation mapping P X|X ∈ F( X|X), the posterior minimizes the cross-entropy metric in (5). In a similar manner, when no privacy constraints are imposed, i.e., when = ∞, and the family P( X|X) is large enough, the Bayesian predictive loss (4) can be exactly characterized as…”

Section: Preliminaries and Fully Collaborative Benchmarkmentioning

confidence: 99%

An Information-Theoretic Analysis of The Cost of Decentralization for Learning and Inference Under Privacy Constraints

Jose,

Simeone

2021

Preprint

View full text Add to dashboard Cite

In vertical federated learning (FL), the features of a data sample are distributed across multiple agents. As such, inter-agent collaboration can be beneficial not only during the learning phase, as is the case for standard horizontal FL, but also during the inference phase. A fundamental theoretical question in this setting is how to quantify the cost, or performance loss, of decentralization for learning and/or inference. In this paper, we consider general supervised learning problems with any number of agents, and provide a novel information-theoretic quantification of the cost of decentralization in the presence of privacy constraints on inter-agent communication within a Bayesian framework. The cost of decentralization for learning and/or inference is shown to be quantified in terms of conditional mutual information terms involving features and label variables.

show abstract

“…Since VFL plays a crucial role in constructing automated decision-making systems in the modern society, it is highly desirable to improve its algorithmic fairness. However, designing fair VFL algorithms is challenging due to two characteristics of VFL [7], [8]. First, the data privacy of all the organizations should be fully protected to secure successful collaborations.…”

Section: Introductionmentioning

confidence: 99%

“…When training a fair model in VFL, enforcing every organization to launch a single local update per communication round results in inefficiency [7], [8].…”

Section: Introductionmentioning

confidence: 99%

Achieving Model Fairness in Vertical Federated Learning

Shi¹,

Fan²,

Zhou³

et al. 2021

Preprint

View full text Add to dashboard Cite

Vertical federated learning (VFL), which enables multiple enterprises possessing non-overlapped features to strengthen their machine learning models without disclosing their private data and model parameters, has received increasing attention lately. Similar to other machine learning algorithms, VFL suffers from fairness issues, i.e., the learned model may be unfairly discriminatory over the group with sensitive attributes. To tackle this problem, we propose a fair VFL framework in this work. First, we systematically formulate the problem of training fair models in VFL, where the learning task is modeled as a constrained optimization problem. To solve it in a federated manner, we consider its equivalent dual form and develop an asynchronous gradient coordinate-descent ascent algorithm, where each data party performs multiple parallelized local updates per communication round to effectively reduce the number of communication rounds. We prove that the algorithm finds a δ-stationary point of the dual objective in O(δ −4 ) communication rounds under mild conditions. Finally, extensive experiments on three benchmark datasets demonstrate the superior performance of our method in training fair models.

show abstract

A Communication Efficient Collaborative Learning Framework for Distributed Features

Cited by 25 publications

References 26 publications

Secure Multi-Party Computation based Privacy Preserving Data Analysis in Healthcare IoT Systems

Secure Multi-Party Computation based Privacy Preserving Data Analysis in Healthcare IoT Systems

An Information-Theoretic Analysis of The Cost of Decentralization for Learning and Inference Under Privacy Constraints

Achieving Model Fairness in Vertical Federated Learning

Contact Info

Product

Resources

About