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

Modeling spatial-temporal interactions among neighboring agents is at theheart of multi-agent problems such as motion forecasting and crowd navigation.Despite notable progress, it remains unclear to which extent modernrepresentations can capture the causal relationships behind agent interactions.In this work, we take an in-depth look at the causal awareness of theserepresentations, from computational formalism to real-world practice. First, wecast doubt on the notion of non-causal robustness studied in the recentCausalAgents benchmark. We show that recent representations are alreadypartially resilient to perturbations of non-causal agents, and yet modelingindirect causal effects involving mediator agents remains challenging. Toaddress this challenge, we introduce a metric learning approach thatregularizes latent representations with causal annotations. Our controlledexperiments show that this approach not only leads to higher degrees of causalawareness but also yields stronger out-of-distribution robustness. To furtheroperationalize it in practice, we propose a sim-to-real causal transfer methodvia cross-domain multi-task learning. Experiments on pedestrian datasets showthat our method can substantially boost generalization, even in the absence ofreal-world causal annotations. We hope our work provides a new perspective onthe challenges and potential pathways towards causally-aware representations ofmulti-agent interactions. Our code is available athttps://github.com/socialcausality.

Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representations