English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Vertical federated learning (VFL) aims to train models from cross-silo datawith different feature spaces stored on different platforms. Existing VFLmethods usually assume all data on each platform can be used for modeltraining. However, due to the intrinsic privacy risks of federated learning,the total amount of involved data may be constrained. In addition, existing VFLstudies usually assume only one platform has task labels and can benefit fromthe collaboration, making it difficult to attract other platforms to join inthe collaborative learning. In this paper, we study the platform collaborationproblem in VFL under privacy constraint. We propose to incent differentplatforms through a reciprocal collaboration, where all platforms can exploitmulti-platform information in the VFL framework to benefit their own tasks.With limited privacy budgets, each platform needs to wisely allocate its dataquotas for collaboration with other platforms. Thereby, they naturally form amulti-party game. There are two core problems in this game, i.e., how toappraise other platforms' data value to compute game rewards and how tooptimize policies to solve the game. To evaluate the contributions of otherplatforms' data, each platform offers a small amount of "deposit" data toparticipate in the VFL. We propose a performance estimation method to predictthe expected model performance when involving different amount combinations ofinter-platform data. To solve the game, we propose a platform negotiationmethod that simulates the bargaining among platforms and locally optimizestheir policies via gradient descent. Extensive experiments on two real-worlddatasets show that our approach can effectively facilitate the collaborativeexploitation of multi-platform data in VFL under privacy restrictions.

Game of Privacy: Towards Better Federated Platform Collaboration under Privacy Restriction