TalkRL: The Reinforcement Learning Podcast cover image

Sai Krishna Gottipati

TalkRL: The Reinforcement Learning Podcast

00:00

How Do You Represent the Chemicals?

Training data shouldn't just be biastite. Hi should also have some useless molecules, so that er, the score can be predicted accurately. How do you represent thehe chemicals? Internally? Sot, the molecules can be represented in different ways. The people who work with smile string o the representan a string converted to the one hot rector and then embeding  and so on. It's based on the experience of how other drugs, how effective past drugs have been. Ye, it thet not necessary drugs, but any kind of molecules.

Play episode from 49:52
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app