Network World Models as Environments for Algorithm Design on Complex Systems

2026-10-01

AI Summary

This study addresses the challenge that algorithm design for complex systems relies on costly Monte Carlo simulations, hindering efficient evaluation of long-term intervention effects. To overcome this, we propose employing action-conditioned world models as rapid evaluators within the algorithm design loop. Specifically, our method leverages diffusion dynamics learning to construct simulation environments encoding agent behaviors. Through full-trajectory rollback and counterfactual probing, it enables action-level credit assignment and surrogate intervention analysis, thereby facilitating iterative algorithm refinement. Experimental results demonstrate that the proposed approach outperforms baselines across eight tasks while achieving a 14.5× speedup in rollback compared to Monte Carlo simulations, significantly overcoming existing computational bottlenecks.

Abstract

World models, which simulate an environment and predict how it changes under actions, are increasingly used in real-world applications such as robotics. Complex systems call for the same tool because the effect of an action is not immediate. Seeding nodes for a campaign, or immunizing nodes against an epidemic, changes little on its own; what matters is the outcome that unfolds over the steps that follow. Designing an algorithm that selects such actions to maximize expected performance on a task is inherently iterative, and every candidate must be scored by the outcome it produces. Obtaining that outcome has relied on simulation, whose cost becomes a bottleneck when candidates are evaluated over many sampled trajectories. We propose an action-conditioned Network World Model that learns a network's diffusion dynamics under interventions over time, applies each action to the network, and predicts the outcome that follows. It serves as a fast evaluator inside an algorithm design loop in which a coding agent designs and refines executable algorithms using feedback from full rollouts, action-level credit, and counterfactual probes over alternative interventions. Across eight network tasks and five diffusion models, the designed algorithms match or exceed the strongest reported baseline in 138 of 141 settings while enabling up to 14.5 times faster rollouts than Monte Carlo simulation. Code will be released upon acceptance.

Problem

• complex systems
• network world models
• algorithm design
• diffusion dynamics
• simulation bottleneck

Innovation

• Network World Models
• Algorithm Design
• Diffusion Dynamics
• Counterfactual Probes
• Coding Agent