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

Adversarial data poisoning is an effective attack against machine learningand threatens model integrity by introducing poisoned data into the trainingdataset. So far, it has been studied mostly for classification, even thoughregression learning is used in many mission critical systems (such as dosage ofmedication, control of cyber-physical systems and managing power supply).Therefore, in the present research, we aim to evaluate all aspects of datapoisoning attacks on regression learning, exceeding previous work both in termsof breadth and depth. We present realistic scenarios in which data poisoningattacks threaten production systems and introduce a novel black-box attack,which is then applied to a real-word medical use-case. As a result, we observethat the mean squared error (MSE) of the regressor increases to 150 percent dueto inserting only two percent of poison samples. Finally, we present a newdefense strategy against the novel and previous attacks and evaluate itthoroughly on 26 datasets. As a result of the conducted experiments, weconclude that the proposed defence strategy effectively mitigates theconsidered attacks.

Data Poisoning Attacks on Regression Learning and Corresponding Defenses