A train--prune--readout--rewrite workflow for interpretable quantum learning

• URL: http://arxiv.org/abs/2609.15139v1
• Date: Mon, 14 Sep 2026 07:12:21 GMT
• Title: A train--prune--readout--rewrite workflow for interpretable quantum learning
• Abstract summary: 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.<n>Here, a train--prune--readout--rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims.<n>The workflow is implemented with complex-valued networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation.
• Score: 5.641464469724376
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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 papers

• Synthics: Synthetic Physics-like Datasets for Machine Learning [0.18275108630751835]
  We generate synthetic regression datasets that structurally resemble physics equations from a given equation corpus.<n>The generated equations match the corpus on all of the eight studied structural features.<n>A gradient-boosted regressor tuned on the synthetic data picks, on average, the 6th-best configuration out of 20 on real data.
  arXiv  Detail & Related papers  (2026-06-04T21:20:08Z)
• On the Mechanism and Dynamics of Modular Addition: Fourier Features, Lottery Ticket, and Grokking [49.1352577985191]
  We present a comprehensive analysis of how two-layer neural networks learn features to solve the modular addition task.<n>Our work provides a full mechanistic interpretation of the learned model and a theoretical explanation of its training dynamics.
  arXiv  Detail & Related papers  (2026-02-18T20:25:13Z)
• The Kinetics of Reasoning: How Chain-of-Thought Shapes Learning in Transformers? [25.29458951592086]
  Chain-of-thought (CoT) supervision can substantially improve transformer performance.<n>We investigate these learning dynamics through the lens of grokking by pretraining transformers on symbolic reasoning tasks.
  arXiv  Detail & Related papers  (2025-10-28T20:14:26Z)
• Provable In-Context Learning of Nonlinear Regression with Transformers [66.99048542127768]
  In-context learning (ICL) is the ability to perform unseen tasks using task specific prompts without updating parameters.<n>Recent research has actively explored the training dynamics behind ICL, with much of the focus on relatively simple tasks.<n>This paper investigates more complex nonlinear regression tasks, aiming to uncover how transformers acquire in-context learning capabilities.
  arXiv  Detail & Related papers  (2025-07-28T00:09:28Z)
• Transformers Are Universally Consistent [14.904264782690639]
  We show that Transformers equipped with softmax-based nonlinear attention are uniformly consistent when tasked with executing Least Squares regression.<n>We derive upper bounds on the empirical error which, in the regime, decay at a provable rate of $mathcalO(t-1/2d)$, where $t$ denotes the number of input tokens and $d$ the embedding dimensionality.
  arXiv  Detail & Related papers  (2025-05-30T12:39:26Z)
• Unsupervised Representation Learning from Sparse Transformation Analysis [79.94858534887801]
  We propose to learn representations from sequence data by factorizing the transformations of the latent variables into sparse components. Input data are first encoded as distributions of latent activations and subsequently transformed using a probability flow model.
  arXiv  Detail & Related papers  (2024-10-07T23:53:25Z)
• In-Context Convergence of Transformers [63.04956160537308]
  We study the learning dynamics of a one-layer transformer with softmax attention trained via gradient descent. For data with imbalanced features, we show that the learning dynamics take a stage-wise convergence process.
  arXiv  Detail & Related papers  (2023-10-08T17:55:33Z)
• From Bricks to Bridges: Product of Invariances to Enhance Latent Space Communication [19.336940758147442]
  It has been observed that representations learned by distinct neural networks conceal structural similarities when the models are trained under similar inductive biases. We introduce a versatile method to directly incorporate a set of invariances into the representations, constructing a product space of invariant components on top of the latent representations. We validate our solution on classification and reconstruction tasks, observing consistent latent similarity and downstream performance improvements in a zero-shot stitching setting.
  arXiv  Detail & Related papers  (2023-10-02T13:55:38Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• A deep learning driven pseudospectral PCE based FFT homogenization algorithm for complex microstructures [68.8204255655161]
  It is shown that the proposed method is able to predict central moments of interest while being magnitudes faster to evaluate than traditional approaches. It is shown, that the proposed method is able to predict central moments of interest while being magnitudes faster to evaluate than traditional approaches.
  arXiv  Detail & Related papers  (2021-10-26T07:02:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.