BI 131 Sri Ramaswamy and Jie Mei: Neuromodulation-aware DNNs
Brain Inspired
00:00
Deep Learning Networks - Bottom Up Versus Top Down?
Top down models are great at reproducing something specific, but they probably can't reproduce something that's a lot more generally. On the other hand, bottom op approach gives us an order to try and understand what level of detail is required to model what phenomenon. And then based on what level of ail is required, i meanyou could just shave off these details and then only focus on what matters to contribute to a certain phenomenon. I think if these two approaches don't come together, i thinkit's a bit of a groping in the dark, right?
Transcript
Play full episode
