124 - Jay McClelland: Deep Learning, Neural Networks, and Artificial Intelligence
Robinson's Podcast
00:00
The Motivation Behind the Interactive Activation Model
When we started with the interactive activation model, I was a very clear-minded empirically oriented scientist who felt that accounting for all aspects of the data was extremely important. Rommel Hart always wanted to illustrate the alternative way of solving the problem and didn't so much care about exactly nailing all the details of the phenomenon. For him, the interactiveactivation model was a way of breaking out of the mold of good old fashioned AI where you stipulate that there are these various structures and you have to decide which one it is that fits the current context. He was always looking for ways of gratedly combining.
Transcript
Play full episode
