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

When interacting with other non-competitive decision-making agents, it iscritical for an autonomous agent to have inferable behavior: Their actions mustconvey their intention and strategy. For example, an autonomous car's strategymust be inferable by the pedestrians interacting with the car. We model theinferability problem using a repeated bimatrix Stackelberg game withobservations where a leader and a follower repeatedly interact. During theinteractions, the leader uses a fixed, potentially mixed strategy. Thefollower, on the other hand, does not know the leader's strategy anddynamically reacts based on observations that are the leader's previousactions. In the setting with observations, the leader may suffer from aninferability loss, i.e., the performance compared to the setting where thefollower has perfect information of the leader's strategy. We show that theinferability loss is upper-bounded by a function of the number of interactionsand the stochasticity level of the leader's strategy, encouraging the use ofinferable strategies with lower stochasticity levels. As a converse result, wealso provide a game where the required number of interactions is lower boundedby a function of the desired inferability loss.

Encouraging Inferable Behavior for Autonomy: Repeated Bimatrix Stackelberg Games with Observations