31: Ian Goodfellow on Inventing GANs with Peter Bauman (Deep Learning Series 02)

Episode page and full transcript: lerandom.art/episodes/ian-goodfellow-on-invent…

In this extra special episode, host Peter Bauman (Monk Antony, Le Random's editor in chief) speaks with prominent AI researcher Ian Goodfellow (lerandom.art/subjects/ian-goodfellow) about the legendary origins of GANs, their unexpected success and indelible impact on both twenty-first-century image making and AI research.

This episode contains Peter and Ian's full conversation and serves as a companion to Monday's written interview, which covered the first half of the discussion only.

Monday's editorial: lerandom.art/editorial/ian-goodfellow-on-inven…

Also in this episode: Alec Radford (lerandom.art/subjects/alec-radford), Mario Klingemann (lerandom.art/subjects/mario-klingemann), NeurIPS (lerandom.art/subjects/neurips), Remi Denton (lerandom.art/subjects/remi-denton), Pablo Picasso (lerandom.art/subjects/pablo-picasso), Andrew Ng (lerandom.art/subjects/andrew-ng), Andy Warhol (lerandom.art/subjects/andy-warhol), Lawrence Lek (lerandom.art/subjects/lawrence-lek), Mat Dryhurst (lerandom.art/subjects/mat-dryhurst)

Read next: Ian Goodfellow on Inventing GANs (lerandom.art/editorial/ian-goodfellow-on-inven…)

Listen next: Ep. 24, Tom White & Gene Kogan on the Birth of GAN Art (Deep Learning Series 01) (lerandom.art/episodes/tom-white-and-gene-kogan…), Ep. 34, Anna Ridler & Sofia Crespo—The Natural History of Machine Learning with Peter Bauman (Deep Learning Series 03) (lerandom.art/episodes/anna-ridler-and-sofia-cr…)

On the Living Timeline: Ian Goodfellow Introduces GANs (2014) (timeline.lerandom.art/m/ian-goodfellow-introdu…)

Chapters 📖:
00:00 Introduction
03:28 What are GANs and how do they work?
06:10 What happened at the Montreal party?
10:49 Did the first GAN really work first try?
19:34 Why did GANs last so long?
21:20 When did LAPGAN and DCGAN take off?
23:52 Is generative modeling the culmination?
26:09 Can GANs be creative or just mimic?
29:28 Who is creative when a GAN is a brush?
37:12 Can an AI be a fully autonomous artist?
41:48 Did GANs lead to text-to-image?
42:18 How did generative models become media?
51:28 Which advances led to StyleGAN?
57:50 Are engineers artists?
1:02:24 Were GANs used as Goodfellow expected?
1:06:48 Did GANs start the push for scale?
1:11:29 Can anyone predict when AGI arrives?
1:19:12 Do all neural nets converge on reality?
1:23:47 Closing Thoughts