Abstract:While Mixture-of-Experts (MoE) architectures define the state-of-the-art, their theoretical success is often attributed to heuristic efficiency rather than geometric expressivity. In this work, we present the first analysis of MoE through the lens of tropical geometry, establishing that the Top-$k$ routing mechanism is algebraically isomorphic to the $k$-th elementary symmetric tropical polynomial. This isomorphism partitions the input space into the Normal Fan of a Hypersimplex, revealing that \textbf{sparsity is combinatorial depth} which scales geometric capacity by the binomial coefficient $\binom{N}{k}$. Moving beyond ambient bounds, we introduce the concept of \textit{Effective Capacity} under the Manifold Hypothesis. We prove that while dense networks suffer from capacity collapse on low-dimensional data, MoE architectures exhibit \textit{Combinatorial Resilience}, maintaining high expressivity via the transversality of routing cones. Translating these theoretical bounds into architectural principles, we derive asymptotic capacity limits for optimal expert granularity and prove that shared experts are geometrically necessary to prevent routing collapse.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2602.03204 [cs.LG] |
| (or arXiv:2602.03204v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2602.03204 arXiv-issued DOI via DataCite |
From: Ye Su [view email]
[v1]
Tue, 3 Feb 2026 07:17:38 UTC (36 KB)
[v2]
Wed, 6 May 2026 08:20:46 UTC (6,296 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。