Episode 32: Jamie Simon, UC Berkeley: On theoretical principles for how neural networks learn and generalize
Generally Intelligent
00:00
How I Got Started in Deep Learning
How did you get into doing deep learning? And also, what initial research questions did you get interested in? Yeah. I joined his lab early in my first year and started working with a graduate student on a project to try to understand optimization landscapes of neural networks. That project was how I got my initial exposure to many of the big ideas. The core feeling that this process, like learning is somehow a process by which the image of the training data is impressed upon the initially random weights of a neural network still feels right to me.
Transcript
Play full episode
