Group-Theoretic Structure Governing Identifiability in Inverse Problems

TL;DR

This paper uses group-representation theory to understand the limits of reconstructing causal structures in symmetric physical systems, and demonstrates how neural networks respecting these symmetries can achieve identifiable causal inference.

Contribution

It formulates inverse causal inference within group-representation theory, revealing fundamental limits and constructing equivariant neural networks that respect these symmetry-based constraints.

Findings

Reconstructability is constrained by SO(3) representation decomposition.

An SO(3)-equivariant neural network reproduces theoretical identifiability limits.

Group-representation structure fundamentally determines causal inverse problem limits.

Abstract

In physical systems possessing symmetry, reconstructing the underlying causal structure from observational data constitutes an inverse problem of fundamental importance. In this work, we formulate the inverse problem of causal inference within the framework of group-representation theory, clarifying the structure of the representation spaces to which the {\it causality} and estimation maps belong. This formulation leads to both theoretical and practical limits of reconstructability (identifiability). We show that the local velocity-gradient tensor, regarded as a {\it causal factor}, can be reconstructed from the orientational motion of suspended particles. In this setting, the estimation map must act as a group homomorphism between the observation and causal spaces, and the reconstructable subspace is constrained by the decomposition structure of the SO(3) representation. Based on this principle, we construct an SO(3)-equivariant neural network (implemented with the e3nn framework) and verify that the identifiability determined by the group-representation structure is reproduced in the actual learning process. These results demonstrate a fundamental principle that the group-representation structure determines the reconstructability (identifiability limit) in inverse problems of causal inference.