Goal-Oriented Reasoning
The discussion delves into the concept of goal regression and the potential advantages of reasoning backward from desired outcomes. Leslie reflects on the early insights of Simon regarding means-ends reasoning, emphasizing how starting from a goal might simplify decision-making. However, she notes that in the AI planning community, this backward reasoning approach has not consistently proven superior, leaving room for further exploration and intuition.In this clip
From this podcast
Lex Fridman Podcast
Leslie Kaelbling: Reinforcement Learning, Planning, and Robotics | Lex Fridman Podcast #15
Related Questions
What are the AI reasoning breakthroughs discussed in the episode Leslie Kaelbling: Reinforcement Learning, Planning, and Robotics | Lex Fridman Podcast #15 and the clip Goal-Oriented Reasoning?
What are the AI reasoning breakthroughs mentioned in the episode Leslie Kaelbling: Reinforcement Learning, Planning, and Robotics | Lex Fridman Podcast #15 and the clip Goal-Oriented Reasoning?
Can AI have complex goals as discussed in the episode Leslie Kaelbling: Reinforcement Learning, Planning, and Robotics | Lex Fridman Podcast #15 and the clip Goal-Oriented Reasoning?