Paperlane

no. 6 · Monday 2 November 2026

Nonlinear Maps Between Irreducible Representations in Graph Neural Networks

Mani Shayestehfar

math.RT · via cs.LG · 18 pp

Abstract

Equivariant linear maps cannot mix non-isomorphic irreducible representations, whereas nonlinear activations can. We develop a degree-resolved version of this nonlinear interaction picture and use the resulting block inventory to construct Specht Triple, a residual on unordered graph triples. On a synthetic structural benchmark with a frozen GIN backbone, Specht Triple reduces aggregate standardised test MSE by 11.05% relative to a parameter-matched Projector Control using the same irreducible projectors and complete linear equivariant space. The results support irreducible decomposition as an inductive bias for nonlinear higher-order equivariant processing.

comments
18 pages; comments are welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.41059 ↗