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

Predicting accurate future trajectories of multiple agents is essential forautonomous systems, but is challenging due to the complex agent interaction andthe uncertainty in each agent's future behavior. Forecasting multi-agenttrajectories requires modeling two key dimensions: (1) time dimension, where wemodel the influence of past agent states over future states; (2) socialdimension, where we model how the state of each agent affects others. Mostprior methods model these two dimensions separately, e.g., first using atemporal model to summarize features over time for each agent independently andthen modeling the interaction of the summarized features with a social model.This approach is suboptimal since independent feature encoding over either thetime or social dimension can result in a loss of information. Instead, we wouldprefer a method that allows an agent's state at one time to directly affectanother agent's state at a future time. To this end, we propose a newTransformer, AgentFormer, that jointly models the time and social dimensions.The model leverages a sequence representation of multi-agent trajectories byflattening trajectory features across time and agents. Since standard attentionoperations disregard the agent identity of each element in the sequence,AgentFormer uses a novel agent-aware attention mechanism that preserves agentidentities by attending to elements of the same agent differently than elementsof other agents. Based on AgentFormer, we propose a stochastic multi-agenttrajectory prediction model that can attend to features of any agent at anyprevious timestep when inferring an agent's future position. The latent intentof all agents is also jointly modeled, allowing the stochasticity in oneagent's behavior to affect other agents. Our method substantially improves thestate of the art on well-established pedestrian and autonomous drivingdatasets.

AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting