A train--prune--readout--rewrite workflow for interpretable quantum learning
Abstract
AI for Science aims not only to predict complex physical systems from data, but also to extract mathematical structure and physically testable representations from learned models. Here, a train--prune--readout--rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims: algebraically equivalent readout of a trained predictor, compact teacher-faithful symbolic rewriting on the sampled physical domain, and transformation-based tests of learned internal representations. The workflow is implemented with complex-valued Kolmogorov--Arnold networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation. In analytically controlled single-qubit tasks, rewriting recovered the quadratic structure of purity, whereas von Neumann entropy yielded only a domain-bounded symbolic surrogate; physics-aligned variable grouping preserved symbolic fidelity. For two-qubit entanglement-related tasks, shared learning exposed a common internal representation whose physical content was interrogated directly. Local-unitary transformations rejected a direct invariant-coordinate interpretation, while fixed-decoder transfer showed that the shared activation carries Pauli-correlation information in a transformation-consistent form. Task-related invariant spectral features were subsequently recovered through low-order nonlinear readouts. Separate predictive tests retained high accuracy for three-qubit classification and controlled ten-qubit purity regression with over one million complex inputs. These results establish an evidence-resolved framework for distinguishing physical grounding, readable computation, faithful symbolic compression and transformation-tested physical structure in constrained complex-valued scientific learning.
Related
- Design and Modeling of the Charge Readout of a SiMOS Quantum Dot with a Single Electron Transistor and CryoCMOS
- Effective Sub-Quantum Readout for Non-Monochromatic Axion Signals in High-Q Haloscopes
- Beyond Binary Rewrite Decisions: Separating Rewrite-Intensity Control from Level Prediction in a Frozen-LLM Generation-and-Selection Pipeline for Korean Easy-Read Drafting
- Near-Quantum-Limited Readout for the Princeton Axion Search using a sub-GHz Kinetic Inductance Traveling Wave Parametric Amplifier
- Problem-Specific Basis Quantum State Readout via Proper Orthogonal Decomposition
BibTeX
@misc{zhang2026trainprunereadoutrewrite,
title = {{A train--prune--readout--rewrite workflow for interpretable quantum learning}},
author = {Siran Zhang and Shuming Cheng and Xiang Li and Jinyi Liu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.15139},
eprint = {2609.15139},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.15139},
}
RIS
TY - UNPB
TI - A train--prune--readout--rewrite workflow for interpretable quantum learning
AU - Siran Zhang
AU - Shuming Cheng
AU - Xiang Li
AU - Jinyi Liu
PY - 2026
DA - 2026/09/14
PB - arXiv
DO - 10.48550/arxiv.2609.15139
UR - https://arxiv.org/abs/2609.15139
AB - AI for Science aims not only to predict complex physical systems from data, but also to extract mathematical structure and physically testable representations from learned models. Here, a train--prune--readout--rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims: algebraically equivalent readout of a trained predictor, compact teacher-faithful symbolic rewriting on the sampled physical domain, and transformation-based tests of learned internal representations. The workflow is implemented with complex-valued Kolmogorov--Arnold networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation. In analytically controlled single-qubit tasks, rewriting recovered the quadratic structure of purity, whereas von Neumann entropy yielded only a domain-bounded symbolic surrogate; physics-aligned variable grouping preserved symbolic fidelity. For two-qubit entanglement-related tasks, shared learning exposed a common internal representation whose physical content was interrogated directly. Local-unitary transformations rejected a direct invariant-coordinate interpretation, while fixed-decoder transfer showed that the shared activation carries Pauli-correlation information in a transformation-consistent form. Task-related invariant spectral features were subsequently recovered through low-order nonlinear readouts. Separate predictive tests retained high accuracy for three-qubit classification and controlled ten-qubit purity regression with over one million complex inputs. These results establish an evidence-resolved framework for distinguishing physical grounding, readable computation, faithful symbolic compression and transformation-tested physical structure in constrained complex-valued scientific learning.
ER -