Proposes the 'Theory Compiler,' a system that automatically translates formal domain specifications into neural architectures with built-in physical or logical constraints.
March 17, 2026
Original Paper
From Specification to Architecture: A Theory Compiler for Knowledge-Guided Machine Learning
arXiv · 2603.14369
The Takeaway
It moves scientific ML from manual, unverified architecture design to a formal compilation process. This ensures that models satisfy domain-specific theories by construction, significantly improving sample efficiency and out-of-distribution generalization compared to standard regularization methods.
From the abstract
Theory-guided machine learning has demonstrated that including authentic domain knowledge directly into model design improves performance, sample efficiency and out-of-distribution generalisation. Yet the process by which a formal domain theory is translated into architectural constraints remains entirely manual, specific to each domain formalism, and devoid of any formal correctness guarantee. This translation is non-transferable between domains, not verified, and does not scale. We propose the