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

Deep Neural Networks (DNNs) have significantly improved the accuracy ofintelligent applications on mobile devices. DNN surgery, which partitions DNNprocessing between mobile devices and multi-access edge computing (MEC)servers, can enable real-time inference despite the computational limitationsof mobile devices. However, DNN surgery faces a critical challenge: determiningthe optimal computing resource demand from the server and the correspondingpartition strategy, while considering both inference latency and MEC serverusage costs. This problem is compounded by two factors: (1) the finitecomputing capacity of the MEC server, which is shared among multiple devices,leading to inter-dependent demands, and (2) the shift in modern DNNarchitecture from chains to directed acyclic graphs (DAGs), which complicatespotential solutions.  In this paper, we introduce a novel Decentralized DNN Surgery (DDS)framework. We formulate the partition strategy as a min-cut and propose aresource allocation game to adaptively schedule the demands of mobile devicesin an MEC environment. We prove the existence of a Nash Equilibrium (NE), anddevelop an iterative algorithm to efficiently reach the NE for each device. Ourextensive experiments demonstrate that DDS can effectively handle varying MECscenarios, achieving up to 1.25$\times$ acceleration compared to thestate-of-the-art algorithm.

Adaptive DNN Surgery for Selfish Inference Acceleration with On-demand Edge Resource