Quantum Generative Adversarial Networks (QGANs)

Quantum Generative Adversarial Networks (QGANs) are an advanced form of generative models that leverage quantum computing to enhance data generation capabilities.

They build on the classical Generative Adversarial Networks (GANs) framework, which consists of two neural networks: a generator and a discriminator:

• Generator: Creates synthetic data that mimics the real training dataset.
• Discriminator: Evaluates the data, distinguishing between real and generated samples.

Why do I want to use QGANs?

• Image Generation: Creating high-resolution images that are indistinguishable from real ones.
• Data Augmentation: Enhancing datasets for training machine learning models.
• Complex Data Analysis: Analyzing intricate data patterns in fields like finance and healthcare.

An initial representation (a map) about my understanding of the main components of QGANs.

The use of hybrid.architecture (Quantum and Classic), makes these generative models accessible to more use cases: for instance in logistics, not everything can be connected in real time to a quantum computer (today).