Local-to-Collective Predictability from Heterogeneous Multi-Agent Interactions

Collective behavior in multi-agent systems emerges from repeated local interactions, yet existing approaches lack an explicit representation of how heterogeneous local interaction changes propagate to collective-level degradation. To address this gap, this work introduces $\text{Interaction Predictability}$ ($\mathrm{IP}$), an agent-level representation of how strongly an agent's spatiotemporal interaction dynamics support prediction of its future motion. Building on this representation, $\text{Local-to-Collective Predictability (L2CP)}$ learns \IP from ego-centric interaction histories and selectively aggregates agent-level predictability through learned relevance into a Collective Predictability Score ($\mathrm{CPS}$). Both representations are learned without using degradation labels or collective-state metrics as training targets. Experiments show that $\mathrm{IP}$ distinguishes heterogeneous local interaction conditions and achieves stronger degraded-agent localization than simple local-statistics baselines, while relevance-weighted $\mathrm{CPS}$ captures collective degradation structure consistently across multiple interaction dynamics. These results establish local interaction predictability as a learned intermediate representation that connects distributed agent-level dynamics to collective degradation.