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

On-device Deep Neural Networks (DNNs) have recently gained more attention dueto the increasing computing power of the mobile devices and the number ofapplications in Computer Vision (CV), Natural Language Processing (NLP), andInternet of Things (IoTs). Unfortunately, the existing efficient convolutionalneural network (CNN) architectures designed for CV tasks are not directlyapplicable to NLP tasks and the tiny Recurrent Neural Network (RNN)architectures have been designed primarily for IoT applications. In NLPapplications, although model compression has seen initial success in on-devicetext classification, there are at least three major challenges yet to beaddressed: adversarial robustness, explainability, and personalization. Here weattempt to tackle these challenges by designing a new training scheme for modelcompression and adversarial robustness, including the optimization of anexplainable feature mapping objective, a knowledge distillation objective, andan adversarially robustness objective. The resulting compressed model ispersonalized using on-device private training data via fine-tuning. We performextensive experiments to compare our approach with both compact RNN (e.g.,FastGRNN) and compressed RNN (e.g., PRADO) architectures in both natural andadversarial NLP test settings.

Adversarially Robust and Explainable Model Compression with On-Device Personalization for Text Classification