Chapter
Clips
The use of modeling and planning as a tool to generate car and robot behavior as well as to model human behavior is discussed as part of the reinforcement learning process, and how reinforcement learning can be applied in either the real world or in simulation.
54:11 - 58:45 (04:34)
Summary
The use of modeling and planning as a tool to generate car and robot behavior as well as to model human behavior is discussed as part of the reinforcement learning process, and how reinforcement learning can be applied in either the real world or in simulation.
ChapterThe Importance of Modeling in AI Development
Episode#81 – Anca Dragan: Human-Robot Interaction and Reward Engineering
PodcastLex Fridman Podcast
The term "out-of-distribution" is a machine learning jargon that describes situations when the data being processed is completely unknown to the system.
58:46 - 1:01:38 (02:52)
Summary
The term "out-of-distribution" is a machine learning jargon that describes situations when the data being processed is completely unknown to the system. In these cases, the algorithm struggles to make accurate predictions, as it lacks data to draw from.
ChapterThe Importance of Modeling in AI Development
Episode#81 – Anca Dragan: Human-Robot Interaction and Reward Engineering
PodcastLex Fridman Podcast
The fear of death is an important concept that needs to be encoded in the common sense reasoning of AI.
1:01:38 - 1:03:21 (01:42)
Summary
The fear of death is an important concept that needs to be encoded in the common sense reasoning of AI. Solving common sense reasoning is essential in achieving the dream of artificial intelligence.