Engagement Process: Rethinking the Temporal Interface of Action and Observation

TL;DR

The paper introduces Engagement Process, a formalism for modeling complex temporal interactions in decision-making, capturing timing issues and enabling richer agent coordination beyond fixed step-based interfaces.

Contribution

It proposes a novel interaction formalism that explicitly models time in action-observation streams, addressing delays and multi-rate coordination in decision processes.

Findings

EP exposes temporal behaviors hidden by step-based interfaces.

EP enables policies to adapt under explicit time costs.

EP supports richer agent-side organization and multi-rate coordination.

Abstract

Task completion in digital and physical environments increasingly involves complex temporal interaction, where actions and observations unfold over different time scales rather than align with fixed observation--action steps. To model such interactions, we propose \emph{Engagement Process} (EP), an interaction formalism that inherits the decision-theoretic structure of POMDPs while making time explicit in the action--observation interface. EP represents actions and observations as decoupled event streams along time, rather than updates paired at fixed decision steps. This interface captures single-agent timing issues such as deliberation latency, delayed feedback, and persistent actions, while supporting richer agent-side organization, multi-rate coordination, and compositional interaction among subsystems. Across toy, LLM-agent, and learning experiments, EP exposes temporal behaviors hidden by step-based interfaces and enables policies to adapt under explicit time costs.