Train Models With Large Data Sets

The self-supervised or supervised with large data sets both help a lot. So we are seeing the equivalent of foundation models in our own way for geometry and kinematics and things like those. Do you want to have anything, John? Yeah, I'll keep it brief. Basically, whenever we train on a larger data set, we see big improvements in our model performance. And basically, whenever we initialize our networks with some pre-training step from some other auxiliary task, we basically see improvements.

Transcript

Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.