QUIVER: Quantum-Informed Views for Enhanced Representations in Large ML Models

2026-06-01

AI Summary

This work addresses a key challenge in classical machine learning: enhancing large models’ capacity to represent high-order statistical structures. It introduces QUIVER, a novel paradigm that, for the first time, incorporates the quantum Fisher information matrix (QFIM) as an intrinsic, basis-independent geometric feature modality into classical learning frameworks. By leveraging variational quantum circuits to extract intrinsic geometric information from data embedded in quantum state manifolds and fusing it with classical neural networks, QUIVER effectively captures high-order correlations that are difficult for conventional methods to model. The approach demonstrates consistent and significant improvements over standard performance metrics on two cross-domain benchmark tasks—molecular property prediction on QM9 and jet classification on JetClass—thereby validating the practical representational gains conferred by quantum geometric features.

Abstract

Large machine learning models benefit substantially from multimodal inputs that provide a complementary view of the same example. We introduce QUIVER (QUantum-Informed Views for Enhanced Representations, a paradigm that enriches classical data-driven features with a quantum Fisher view: a geometrically motivated, basis-independent summary of higher-order correlations captured by a variational quantum circuit (VQC) trained to perform the same task. Unlike classical feature augmentation, the quantum Fisher information matrix encodes the intrinsic geometry of the learned quantum state manifold. While this feature map, motivated by quantum information theory, is ordinarily non-trivial to model classically, it can surface statistical structure that additional classical data or model capacity finds difficult to learn. This makes the quantum Fisher view a genuinely complementary modality rather than a redundant one. We demonstrate that QUIVER improves standard performance metrics on two benchmark datasets from very different fields: QM9 for predicting molecule properties, and JetClass for predicting jet flavor at the Large Hadron Collider (LHC). The core contribution, however, is domain-agnostic: the quantum Fisher view can be fused into a broad class of model architectures via targeted modifications to the base architecture, to incorporate information about the quantum geometry of the problem. These results demonstrate that quantum-geometric features, extracted from simulated variational circuits, can deliver measurable value for standard machine learning tasks, well before the advent of fault-tolerant quantum hardware.

Problem

• quantum Fisher information
• feature augmentation
• multimodal learning
• variational quantum circuits
• quantum geometry

Innovation

• quantum Fisher information
• variational quantum circuit
• geometric feature augmentation
• multimodal representation learning
• quantum-classical hybrid models