Cathedral
New researchDiscoverProjectsLibraryAgentsTools
ActivitySettings

Evidence, people, agents, and tools in one research record.

Cathedral
AskDiscoverProjectsLibrary
Cathedral
New researchDiscoverProjectsLibraryAgentsTools
ActivitySettings

Evidence, people, agents, and tools in one research record.

Cathedral

Paper / Europe Pmc

On the generalization limits of quantum generative adversarial networks with pure state generators.

We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks. Our analysis centers on fully quantum implementations of both the generator and discriminator. Through extensive numerical testing of current main architectures, we find that QGANs struggle to generalize across datasets, converging on merely the average representation of the training data. When the output of the generator is a pure-state

Frkatovic JMalemath AKankeu IWerner YTschöpe MRey VFSuh SLukowicz PPalaiodimopoulos NKiefer-Emmanouilidis M.

We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks. Our analysis centers on fully quantum implementations of both the generator and discriminator. Through extensive numerical testing of current main architectures, we find that QGANs struggle to generalize across datasets, converging on merely the average representation of the training data.

When the output of the generator is a pure-state, we analytically derive a lower bound for the discriminator quality given by the fidelity between the pure-state output of the generator and the target data distribution, thereby providing a theoretical explanation for the limitations observed in current models. Our findings reveal fundamental challenges in the generalization capabilities of existing quantum generative models. While our analysis focuses on QGANs, the results carry broader implications for the performance of related quantum generative models.

2026-06-09

Research

MED:42265136

Evidence studio

Auditable passages, figures, and tables

No quote evidence has been extracted yet. Cathedral will preserve page references and keep quoted passages short enough to verify quickly.
A full PDF is not available from this source. Abstract-level evidence remains available inside Cathedral.

Layered summaries

eli5

This paper studies on the generalization limits of quantum generative adversarial networks with pure state generators. and argues: We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks.

standard

We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks. Our analysis centers on fully quantum implementations of both the generator and discriminator. Through extensive numerical testing of current main architectures, we find that QGANs struggle to generalize across datasets, converging on merely the average representation of the training data. When the output of the generator is a pure-state

expert

We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks. Our analysis centers on fully quantum implementations of both the generator and discriminator. Through extensive numerical testing of current main architectures, we find that QGANs struggle to generalize across datasets, converging on merely the average representation of the training data. When the output of the generator is a pure-state Source: europe-pmc (10.1038/s41598-026-54900-7)

Public review

Publish a transparent judgment. Scores are signals, not a substitute for the written rationale.

AskDiscoverProjectsLibrary

Replication registry

Extracted claims

We investigate the capabilities of quantum generative adversarial networks (QGANs) in image generations tasks.

confidence4.0

Assumptions

Assumption extraction is not loaded for this paper yet.

Review signals

fraud risk0.0
claim density2.5
assumption coverage0.0

Actions

Back to researchOriginal record

Progress ledger

The ledger will record evidence extraction, reviews, replications, and accepted figures.