Skip to results
MLSift

Titles, abstracts, or an arXiv ID

← Back to results
routineTheory & OptimizationSusceptibility Analysis2609.04661

Interpretability for Turing Machines

Billy Snikkers, Rumi Salazar, Daniel Murfet, Will Troiani

cs.LG cs.FL stat.ML

Abstract

We show that susceptibilities, an interpretability technique developed for neural networks, can identify the presence of algorithmic structure in Turing machines by probing the local loss landscape of a learning problem for noisy Turing machines introduced by Murfet and Troiani (arXiv:2504.08075). We prove that symmetries and path separation in the algorithm implemented by a Turing machine induce permutation symmetries and low-rank blocks in its susceptibility matrix. We study this empirically on a set of deterministic finite automata (DFAs) and demonstrate that algorithmic features can be recovered by principal component analysis and clustering methods in susceptibility space.

Topics

Classified with taxonomy v2 on Mon, 7 Sept 2026.

Report a classification error

Loading the PDF downloads the document. Open it in your browser's viewer, or load it here.

Open PDF