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

Stragglers, Byzantine workers, and data privacy are the main bottlenecks indistributed cloud computing. Some prior works proposed coded computingstrategies to jointly address all three challenges. They require either a largenumber of workers, a significant communication cost or a significantcomputational complexity to tolerate Byzantine workers. Much of the overhead inprior schemes comes from the fact that they tightly couple coding for all threeproblems into a single framework. In this paper, we propose Adaptive VerifiableCoded Computing (AVCC) framework that decouples the Byzantine node detectionchallenge from the straggler tolerance. AVCC leverages coded computing just forhandling stragglers and privacy, and then uses an orthogonal approach thatleverages verifiable computing to mitigate Byzantine workers. Furthermore, AVCCdynamically adapts its coding scheme to trade-off straggler tolerance withByzantine protection. We evaluate AVCC on a compute-intensive distributedlogistic regression application. Our experiments show that AVCC achieves up to$4.2\times$ speedup and up to $5.1\%$ accuracy improvement over thestate-of-the-art Lagrange coded computing approach (LCC). AVCC also speeds upthe conventional uncoded implementation of distributed logistic regression byup to $7.6\times$, and improves the test accuracy by up to $12.1\%$.

Adaptive Verifiable Coded Computing: Towards Fast, Secure and Private Distributed Machine Learning