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

Machine learning models based on Deep Neural Networks (DNNs) are increasinglydeployed in a wide range of applications ranging from self-driving cars toCOVID-19 treatment discovery. To support the computational power necessary tolearn a DNN, cloud environments with dedicated hardware support have emerged ascritical infrastructure. However, there are many integrity challengesassociated with outsourcing computation. Various approaches have been developedto address these challenges, building on trusted execution environments (TEE).Yet, no existing approach scales up to support realistic integrity-preservingDNN model training for heavy workloads (deep architectures and millions oftraining examples) without sustaining a significant performance hit. Tomitigate the time gap between pure TEE (full integrity) and pure GPU (nointegrity), we combine random verification of selected computation steps withsystematic adjustments of DNN hyper-parameters (e.g., a narrow gradientclipping range), hence limiting the attacker's ability to shift the modelparameters significantly provided that the step is not selected forverification during its training phase. Experimental results show the newapproach achieves 2X to 20X performance improvement over pure TEE basedsolution while guaranteeing a very high probability of integrity (e.g., 0.999)with respect to state-of-the-art DNN backdoor attacks.

GOAT: GPU Outsourcing of Deep Learning Training With Asynchronous Probabilistic Integrity Verification Inside Trusted Execution Environment