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

The recent years have been marked by extended research on adversarialattacks, especially on deep neural networks. With this work we intend on posingand investigating the question of whether the phenomenon might be more generalin nature, that is, adversarial-style attacks outside classical classificationtasks. Specifically, we investigate optimization problems as they constitute afundamental part of modern AI research. To this end, we consider the base classof optimizers namely Linear Programs (LPs). On our initial attempt of a na\"ivemapping between the formalism of adversarial examples and LPs, we quicklyidentify the key ingredients missing for making sense of a reasonable notion ofadversarial examples for LPs. Intriguingly, the formalism of Pearl's notion tocausality allows for the right description of adversarial like examples forLPs. Characteristically, we show the direct influence of the Structural CausalModel (SCM) onto the subsequent LP optimization, which ultimately exposes anotion of confounding in LPs (inherited by said SCM) that allows foradversarial-style attacks. We provide both the general proof formally alongsideexistential proofs of such intriguing LP-parameterizations based on SCM forthree combinatorial problems, namely Linear Assignment, Shortest Path and areal world problem of energy systems.

Structural Causal Models Reveal Confounder Bias in Linear Program Modelling