Quantum Scrambling Born Machine

2026-02-19

AI Summary

This work proposes a novel quantum generative model to address the high training complexity of noisy intermediate-scale quantum devices in generative modeling. The model employs a decoupled architecture wherein a fixed multi-body entangling unit—referred to as a “scrambling pool”—provides global entanglement, while only single-qubit rotation parameters are optimized. The generative task is recast as a variational Hamiltonian problem. By integrating Born machines, Haar-random unitaries, brick-wall circuits, and spin-chain Hamiltonian evolution, the approach substantially reduces the number of trainable parameters. Experimental results demonstrate that the model’s performance is largely insensitive to the specific implementation of the scrambler and achieves representational capacity comparable to classical generative models at equivalent parameter counts.

Technology Category

• Machine Learning: Quantum Machine Learning
• Natural Language Processing: Generation
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Quantum generative modeling, where the Born rule naturally defines probability distributions through measurement of parameterized quantum states, is a promising near-term application of quantum computing. We propose a Quantum Scrambling Born Machine in which a fixed entangling unitary -- acting as a scrambling reservoir -- provides multi-qubit entanglement, while only single-qubit rotations are optimized. We consider three entangling unitaries -- a Haar random unitary and two physically realizable approximations, a finite-depth brickwork random circuit and analog time evolution under nearest-neighbor spin-chain Hamiltonians -- and show that, for the benchmark distributions and system sizes considered, once the entangler produces near-Haar-typical entanglement the model learns the target distribution with weak sensitivity to the scrambler's microscopic origin. Finally, promoting the Hamiltonian couplings to trainable parameters casts the generative task as a variational Hamiltonian problem, with performance competitive with representative classical generative models at matched parameter count.

Problem

• quantum generative modeling
• quantum scrambling
• Born machine
• entangling unitary
• variational Hamiltonian

Innovation

• Quantum Scrambling
• Born Machine
• Entangling Unitary
• Variational Hamiltonian
• Quantum Generative Modeling