Chapter
The Limitations of Linguistic Approaches to Drug Development
46:14 - 55:38 (09:24)
The speaker discusses how linguistic approaches to drug development have been limited in their successes compared to models that utilize other sources such as domain experts and combinatorial spaces of molecules or graphs.
Clips
This podcast discusses the use of AI in drug design and how it can improve the drug discovery process by allowing chemists to analyze and understand millions of molecules and reactions that may not be possible for them to do manually.
46:14 - 50:45 (04:31)
Summary
This podcast discusses the use of AI in drug design and how it can improve the drug discovery process by allowing chemists to analyze and understand millions of molecules and reactions that may not be possible for them to do manually.
ChapterThe Limitations of Linguistic Approaches to Drug Development
EpisodeRegina Barzilay: Deep Learning for Cancer Diagnosis and Treatment
PodcastLex Fridman Podcast
The speaker discusses the evolution of natural language processing techniques, starting from rule-based approaches to the current corpus-based techniques that utilize statistics and prediction models, with fewer references to linguistics.
50:45 - 52:36 (01:50)
Summary
The speaker discusses the evolution of natural language processing techniques, starting from rule-based approaches to the current corpus-based techniques that utilize statistics and prediction models, with fewer references to linguistics.
ChapterThe Limitations of Linguistic Approaches to Drug Development
EpisodeRegina Barzilay: Deep Learning for Cancer Diagnosis and Treatment
PodcastLex Fridman Podcast
The use of syntactic trees in linguistics to represent language is limited and not fully understandable by humans, which hinders significant progress in machine translation systems.
52:36 - 55:38 (03:02)
Summary
The use of syntactic trees in linguistics to represent language is limited and not fully understandable by humans, which hinders significant progress in machine translation systems.