Bridging Local and Global Knowledge: Cascaded Mixture-of-Experts Learning for Near-Shortest Path Routing

2026-03-16

AI Summary

This work addresses the limited generalization of deep learning models in sparse networks due to insufficient global structural awareness, which degrades near-shortest-path routing performance. To overcome this, the authors propose a Cascaded Mixture-of-Experts (Ca-MoE) architecture that enables collaborative decision-making between bottom-layer local experts and top-layer global experts. An adaptive inference mechanism activates upper-layer experts only when necessary, balancing computational efficiency with representational capacity. By integrating online meta-learning with stability-oriented parameter updates, the method effectively fuses local and global knowledge while mitigating catastrophic forgetting. Experimental results demonstrate that the approach improves routing accuracy by up to 29.1% over single-expert baselines in sparse networks and consistently maintains performance within 1%–6% of the theoretical optimum across varying graph densities.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

While deep learning models that leverage local features have demonstrated significant potential for near-optimal routing in dense Euclidean graphs, they struggle to generalize well in sparse networks where topological irregularities require broader structural awareness. To address this limitation, we train a Cascaded Mixture of Experts (Ca-MoE) to solve the all-pairs near-shortest path (APNSP) routing problem. Our Ca-MoE is a modular two-tier architecture that supports the decision-making for forwarder selection with lower-tier experts relying on local features and upper-tier experts relying on global features. It performs adaptive inference wherein the upper-tier experts are triggered only when the lower-tier ones do not suffice to achieve adequate decision quality. Computational efficiency is thus achieved by escalating model capacity only when necessitated by topological complexity, and parameter redundancy is avoided. Furthermore, we incorporate an online meta-learning strategy that facilitates independent expert fine-tuning and utilizes a stability-focused update mechanism to prevent catastrophic forgetting as new graph environments are encountered. Experimental evaluations demonstrate that Ca-MoE routing improves accuracy by up to 29.1% in sparse networks compared to single-expert baselines and maintains performance within 1%-6% of the theoretical upper bound across diverse graph densities.

Problem

• near-shortest path routing
• sparse networks
• topological irregularities
• generalization
• all-pairs near-shortest path

Innovation

• Cascaded Mixture-of-Experts
• adaptive inference
• global-local knowledge integration
• online meta-learning
• near-shortest path routing