Latent Space: The AI Engineer Podcast cover image

ICLR 2024 — Best Papers & Talks (ImageGen, Vision, Transformers, State Space Models) ft. Durk Kingma, Christian Szegedy, Ilya Sutskever

Latent Space: The AI Engineer Podcast

NOTE

Understanding and manipulating internal model presentations

The model's neglect of certain subjects in the input font led to the use of internal presentations and tension maps as a direct loss to extract explanations. By understanding the internal presentations of the model, one can manipulate and control it to achieve desired outcomes. Learning representations as compression, including unsupervised learning, ties into auto-encoders and generative models, as discussed by Ilya Sutsgever.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner