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

Recent studies have revealed that neural network-based policies can be easilyfooled by adversarial examples. However, while most prior works analyze theeffects of perturbing every pixel of every frame assuming white-box policyaccess, in this paper we take a more restrictive view towards adversarygeneration - with the goal of unveiling the limits of a model's vulnerability.In particular, we explore minimalistic attacks by defining three key settings:(1) black-box policy access: where the attacker only has access to the input(state) and output (action probability) of an RL policy; (2) fractional-stateadversary: where only several pixels are perturbed, with the extreme case beinga single-pixel adversary; and (3) tactically-chanced attack: where onlysignificant frames are tactically chosen to be attacked. We formulate theadversarial attack by accommodating the three key settings and explore theirpotency on six Atari games by examining four fully trained state-of-the-artpolicies. In Breakout, for example, we surprisingly find that: (i) all policiesshowcase significant performance degradation by merely modifying 0.01% of theinput state, and (ii) the policy trained by DQN is totally deceived byperturbation to only 1% frames.

Minimalistic Attacks: How Little it Takes to Fool a Deep Reinforcement Learning Policy