Latent Style-based Quantum Wasserstein GAN for Drug Design
quant-ph, cs.AI, cs.LG, q-bio.BM
Submitted: 2026-03-23
Updated: 2026-10-28
Comments: Main part: 26 pages, 11 figures, 6 tables. Supplementary material: 24 pages, 20 figures, 21 tables. v2: revision containing matched-capacity analysis of the classical GAN and extensive details on the hardware inference run, plus bootstrap error analysis of the hardware results. Matches published version
Journal ref: J. Chem. Inf. Model. (2026) 66 (20): 13136-13154
License: http://creativecommons.org/licenses/by/4.0/
The gist: The development of new drugs is a time-consuming and expensive process, with average costs estimated to be up to around 2.5 billion.
Terminology
Abstract
The development of new drugs is a time-consuming and expensive process, with average costs estimated to be up to around 2.5 billion. The first step in this process is the design of the new drug, for which de novo drug design, assisted by artificial intelligence, has blossomed in recent years and has revolutionized the field. In particular, generative artificial intelligence has delivered promising results in drug discovery and development, reducing costs and time to solution. However, classical generative models, such as generative adversarial networks (GANs), can be difficult to train and are prone to mode collapse. Quantum computing may be an avenue to overcome some of these issues, providing compact models with far fewer trainable parameters, thereby enhancing the generalizability of GANs. We propose a new latent style-based quantum Wasserstein GAN (QGAN) for drug design, combining a variational autoencoder that maps molecular structures into a latent space with a hybrid quantum GAN. Our pipeline is benchmarked against classical models of the industry-relevant MOSES suite. Style-based noise reuploading at every rotational gate and a Wasserstein gradient penalty are used to improve trainability and to mitigate mode collapse. Training and large-scale sampling with circuits of up to 15 qubits are performed on noiseless quantum simulators to validate our architecture. We present an inference-only demonstration on an IBM Heron quantum computer using 5 qubits, sampling 2,500 new molecules, to investigate quantum noise effects. Our default QGAN, using 110 trainable generator parameters, reaches statistically comparable overall performance and statistically significant gains on selected drug-relevant metrics relative to both the MOSES classical latent GAN baseline, which contains about 700,000 trainable parameters for the generator, and a classical latent GAN with matched generator capacity.
Sources
- Sample-based training of quantum generative models
- The Spectral Amplitude Principle for Dynamics of Quantum Neural Networks
- Spectral Bias in Variational Quantum Machine Learning
- MolGAN: An implicit generative model for small molecular graphs
- Cross-platform hardware benchmark of style-based quantum GANs for data augmentation on superconducting and trapped-ion processors
- Latent Style-based Quantum GAN for high-quality Image Generation
- Exponential capacity scaling of classical GANs compared to hybrid latent style-based quantum GANs
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- PennyLane: Automatic differentiation of hybrid quantum-classical computations
Related papers
- Computational regimes in matrix-product-state-based quantum trajectory simulations
- Toward chemically accurate quantum simulations on quantum hardware: A quantum-HPC hybrid approach
- Transversal surface-code game with reconfigurable qubits
- Fulqrum: a generalized framework for quantum subspace diagonalization
- Flux-Activated Resonant Control of a Bosonic Quantum Memory
- Machine learning the effects of many quantum measurements