Episode Details
Back to Episodes
Deep Dive: Learning Latent Action World Models In The Wild
Published 6 months, 3 weeks ago
Description
An in-depth exploration of "Learning Latent Action World Models In The Wild" - a paper that presents novel approaches to world modeling for reinforcement learning agents. We discuss how agents can learn to understand and predict the dynamics of complex environments without explicit action labels, the technical innovations behind this approach, and implications for building more capable AI systems that can learn from unlabeled video data.
Paper: https://arxiv.org/abs/2501.04045
This podcast is from Colin Davis (colin-davis.com) using Claude & Elevenlabs.