Model-Based Learning
Chelsea discusses the distinction between learned models and known models in robotics, emphasizing the importance of model-based control. She explains how predictions can be made not just in pixel space but through latent representations, allowing for a more efficient approach in reinforcement learning tasks. The conversation highlights the relevance of these methods in real-world applications where models are not predefined.In this clip
From this podcast
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Trends in Reinforcement Learning with Chelsea Finn - #335
Related Questions
As we have robots interact in the physical world, is that a signal that could be used in reinforcement learning in the context of the episode Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10 and the clip Hierarchical Reasoning Challenges?
So as we have robots interact in the physical world, is that a signal that could be used in reinforcement learning in the context of the episode Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10 and the clip Robot Psychology?
As we have robots interact in the physical world, is that a signal that could be used in reinforcement learning in the context of the episode Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10 and the clip Robot Psychology?