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

Reward function, as an incentive representation that recognizes humans'agency and rationalizes humans' actions, is particularly appealing for modelinghuman behavior in human-robot interaction. Inverse Reinforcement Learning is aneffective way to retrieve reward functions from demonstrations. However, it hasalways been challenging when applying it to multi-agent settings since themutual influence between agents has to be appropriately modeled. To tackle thischallenge, previous work either exploits equilibrium solution concepts byassuming humans as perfectly rational optimizers with unbounded intelligence orpre-assigns humans' interaction strategies a priori. In this work, we advocatethat humans are bounded rational and have different intelligence levels whenreasoning about others' decision-making process, and such an inherent andlatent characteristic should be accounted for in reward learning algorithms.Hence, we exploit such insights from Theory-of-Mind and propose a newmulti-agent Inverse Reinforcement Learning framework that reasons about humans'latent intelligence levels during learning. We validate our approach in bothzero-sum and general-sum games with synthetic agents and illustrate a practicalapplication to learning human drivers' reward functions from real driving data.We compare our approach with two baseline algorithms. The results show that byreasoning about humans' latent intelligence levels, the proposed approach hasmore flexibility and capability to retrieve reward functions that explainhumans' driving behaviors better.

Learning Human Rewards by Inferring Their Latent Intelligence Levels in Multi-Agent Games: A Theory-of-Mind Approach with Application to Driving Data