S2E1: A Conversation with the Broad Institute's Anne Carpenter
Theory and Practice
The Power of Feature Learning in Natural Language
The first paper that really proved this out was a technique called word to vac. This is ancient history, there's been many follow up works on top of it. What they were doing was turning words into vectors as opposed to images into vector. And what they found is that not only were these featureizations good for making predictions, but also could do arithmetic with them.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
