Common Sense Limitations
Daniel and Yejin discuss the limitations of neural networks in understanding common sense concepts, particularly in areas related to time and numeracy. They explore the need for better representation of these concepts and the potential of neural networks in the future.In this clip
From this podcast
The Gradient
Yejin Choi: Teaching Machines Common Sense and Morality
Related Questions
Can neural networks be made to reason as discussed in the episode Social Commonsense Reasoning with Yejin Choi - 518 and the clip Reasoning in AI, as well as in the episode Ilya Sutskever: Deep Learning | Lex Fridman Podcast #94 and the clip Neural Networks and Reasoning?
How does AI learn common sense in the context of the episode Joscha Bach: Artificial Consciousness and the Nature of Reality | Lex Fridman Podcast #101 and the clip Common Sense Reasoning?