Tarun Chitra: Drinking 3 Redbulls a Day, Gauntlet’s Financial Modelling Platform, and the Future of Governance
The Delphi Podcast
00:00
Is It That Self Driving Cars Take Up Too Much Memory or Compute?
The amount of compute needed tok train a model is declining. I don't think that this idea like permanently static models is going t rand thats that's why self driving cars are just like, you know, they were, like the hype cycle in two thousand 15 and like, they have not done super well. We do have this weird society l double standard. And i do think that things like that change how you build a model. You went from a model that needs to be good on average to a model it neds to be good but that's very hard for these nerlnet type of things to do. They're much better at average case and like, near average case perform
Transcript
Play full episode
