Episode Notes

Host Nono Martínez Alonso and Aziz Barbar on StyleGAN, NVIDIA's state-of-the-art machine learning algorithm that generates convincing images.

Listen to this episode to learn about StyleGAN.

Nono Martínez Alonso hosts the Getting Simple podcast, sketches things that call his attention, writes stories about enjoying a slower life, and records live streams and tutorials on creative coding and machine intelligence.

Aziz Barbar is an architect and design technologist. He occasionally teaches design courses on computation and the built environment.

Favorite quotes

• "Do we learn less because we can skip certain steps?"
• "Do we skip the process of hiring a designer if we have a computer algorithm that can generate those paintings for us?"
• "StyleGAN is an algorithm open-sourced by NVIDIA - that's the company that creates the graphic cards - that can learn from the features or style of a bunch of images and generate synthetic (or fake) images that resemble the original ones."

Links

• StyleGAN
• Pix2Pix
• Generative adversarial network
• Machine learning
• Image classification
• Latent space

People mentioned

• Ian Goodfellow
• Phillip Isola

Chapters

• 0:00 · Intro
• 0:28 · What is StyleGAN?
• 1:39 · Adversarial networks
• 4:03 · Unexpectedness
• 5:44 · Does it know what it's drawing?
• 7:37 · Latent space
• 8:42 · Hallucinations
• 9:37 · Creative block

I'd love to hear from you.

Submit a question about this or any previous episodes.

Join the Discord community. Meet other curious minds.

If you enjoy the show, would you please consider leaving a short review on Apple Podcasts/iTunes? It takes less than 60 seconds and really helps.

When you buy through links on Getting Simple, I may earn an affiliate commission.

The cover image is a derivative from NVIDIA's RTX 2080 graphics card.