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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Federated learning (FL) is a common and practical framework for learning amachine model in a decentralized fashion. A primary motivation behind thisdecentralized approach is data privacy, ensuring that the learner never seesthe data of each local source itself. Federated learning then comes with twomajors challenges: one is handling potentially complex model updates between aserver and a large number of data sources; the other is that de-centralizationmay, in fact, be insufficient for privacy, as the local updates themselves canreveal information about the sources' data. To address these issues, weconsider an approach to federated learning that combines quantization anddifferential privacy. Absent privacy, Federated Learning often relies onquantization to reduce communication complexity. We build upon this approachand develop a new algorithm called the \textbf{R}andomized\textbf{Q}uantization \textbf{M}echanism (RQM), which obtains privacy through atwo-levels of randomization. More precisely, we randomly sub-sample feasiblequantization levels, then employ a randomized rounding procedure using thesesub-sampled discrete levels. We are able to establish that our results preserve``Renyi differential privacy'' (Renyi DP). We empirically study the performanceof our algorithm and demonstrate that compared to previous work it yieldsimproved privacy-accuracy trade-offs for DP federated learning. To the best ofour knowledge, this is the first study that solely relies on randomizedquantization without incorporating explicit discrete noise to achieve Renyi DPguarantees in Federated Learning systems.

Randomized Quantization is All You Need for Differential Privacy in Federated Learning