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

We consider the problem of prediction by a machine learning algorithm, calledlearner, within an adversarial learning setting. The learner's task is tocorrectly predict the class of data passed to it as a query. However, alongwith queries containing clean data, the learner could also receive malicious oradversarial queries from an adversary. The objective of the adversary is toevade the learner's prediction mechanism by sending adversarial queries thatresult in erroneous class prediction by the learner, while the learner'sobjective is to reduce the incorrect prediction of these adversarial querieswithout degrading the prediction quality of clean queries. We propose a gametheory-based technique called a Repeated Bayesian Sequential Game where thelearner interacts repeatedly with a model of the adversary using self play todetermine the distribution of adversarial versus clean queries. It thenstrategically selects a classifier from a set of pre-trained classifiers thatbalances the likelihood of correct prediction for the query along with reducingthe costs to use the classifier. We have evaluated our proposed technique usingclean and adversarial text data with deep neural network-based classifiers andshown that the learner can select an appropriate classifier that iscommensurate with the query type (clean or adversarial) while remaining awareof the cost to use the classifier.

Playing to Learn Better: Repeated Games for Adversarial Learning with Multiple Classifiers