English

Certifiable Boolean Reasoning Is Universal

Computational Complexity 2026-02-06 v1 Machine Learning

Abstract

The proliferation of agentic systems has thrust the reasoning capabilities of AI into the forefront of contemporary machine learning. While it is known that there \emph{exist} neural networks which can reason through any Boolean task f:{0,1}B{0,1}f:\{0,1\}^B\to\{0,1\}, in the sense that they emulate Boolean circuits with fan-in 22 and fan-out 11 gates, trained models have been repeatedly demonstrated to fall short of these theoretical ideals. This raises the question: \textit{Can one exhibit a deep learning model which \textbf{certifiably} always reasons and can \textbf{universally} reason through any Boolean task?} Moreover, such a model should ideally require few parameters to solve simple Boolean tasks. We answer this question affirmatively by exhibiting a deep learning architecture which parameterizes distributions over Boolean circuits with the guarantee that, for every parameter configuration, a sample is almost surely a valid Boolean circuit (and hence admits an intrinsic circuit-level certificate). We then prove a universality theorem: for any Boolean f:{0,1}B{0,1}f:\{0,1\}^B\to\{0,1\}, there exists a parameter configuration under which the sampled circuit computes ff with arbitrarily high probability. When ff is an O(logB)\mathcal{O}(\log B)-junta, the required number of parameters scales linearly with the input dimension BB. Empirically, on a controlled truth-table completion benchmark aligned with our setting, the proposed architecture trains reliably and achieves high exact-match accuracy while preserving the predicted structure: every internal unit is Boolean-valued on {0,1}B\{0,1\}^B. Matched MLP baselines reach comparable accuracy, but only about 10%10\% of hidden units admit a Boolean representation; i.e.\ are two-valued over the Boolean cube.

Keywords

Cite

@article{arxiv.2602.05120,
  title  = {Certifiable Boolean Reasoning Is Universal},
  author = {Wenhao Li and Anastasis Kratsios and Hrad Ghoukasian and Dennis Zvigelsky},
  journal= {arXiv preprint arXiv:2602.05120},
  year   = {2026}
}

Comments

Submitted to COLT 2026