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

This article seeks for a distributed learning solution for the visualtransformer (ViT) architectures. Compared to convolutional neural network (CNN)architectures, ViTs often have larger model sizes, and are computationallyexpensive, making federated learning (FL) ill-suited. Split learning (SL) candetour this problem by splitting a model and communicating the hiddenrepresentations at the split-layer, also known as smashed data.Notwithstanding, the smashed data of ViT are as large as and as similar as theinput data, negating the communication efficiency of SL while violating dataprivacy. To resolve these issues, we propose a new form of CutSmashed data byrandomly punching and compressing the original smashed data. Leveraging this,we develop a novel SL framework for ViT, coined CutMixSL, communicatingCutSmashed data. CutMixSL not only reduces communication costs and privacyleakage, but also inherently involves the CutMix data augmentation, improvingaccuracy and scalability. Simulations corroborate that CutMixSL outperformsbaselines such as parallelized SL and SplitFed that integrates FL with SL.

Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning