
Episode 07: Yujia Huang, Caltech, on neuro-inspired generative models
Generally Intelligent
00:00
How to Take Good Things From Neuroscience
Backprop is way more efficient than any biologically plausible optimization. It's possible that we actually have a better optimizer than our brain does maybe, and we're still able to get to the same place. I've been thinking about how to properly conduct research that take inspirations from neuroscience. As we find out more about brain, I think we'll keep uncovering more things with its parts.
Transcript
Play full episode