Policy-evaluation and capability-selection complexity

Determine the computational cost of policy evaluation as a function of policy-set size, candidate-set size, and policy structural complexity, and characterize how candidate-generation complexity grows with capability-registry size.

Background

The paper gives worst-case asymptotic estimates for candidate generation and policy filtering but does not measure their constant factors or practical operating regimes. Such measurements are needed to determine when caching, incremental evaluation, registry sharding, policy indexing, or tiered evaluation become necessary.

This problem concerns both the policy stage and the capability-discovery stage, whose costs may dominate the decision cycle as deployments scale.

References

EQ6 --- Policy evaluation overhead and capability selection complexity. What is the computational cost of policy evaluation as a function of policy set size $|P|$, candidate set size $|C'|$, and the structural complexity of individual policies (e.g., conjunctive depth, number of context attributes referenced)? Similarly, how does candidate generation complexity grow with registry size $|C|$? In the worst case, candidate generation is $O(|C|)$ and policy filtering is $O(|C| \cdot |P|)$; empirical evaluation should characterize constant factors and identify the regime at which caching or incremental evaluation becomes necessary. These measurements inform practical deployment decisions about registry sharding, policy indexing, and tiered evaluation strategies.

— Brain API: An Intent-Aware Control Plane for Policy-Governed Agentic Systems  (2609.21299 - Chernov, 18 Sep 2026) in Section 12.7, Evaluation Criteria (EQ6)