Disclaimer: I work in the lab that created Spaun. This seems kind of similar to Spaun [0]. You use Deep Learning to get the features you want and then you do s

Disclaimer: I work in the lab that created Spaun. This seems kind of similar to Spaun [0]. You use Deep Learning to get the features you want and then you do symbolic like operations on those feature vectors to accomplish reinforcement learning. At least Spaun used neurons for the rule manipulation too. I also don't believe the transfer learning claim at all, because the tasks are way too similar. Still, it's a step in the right direction as far as I'm concerned [1]. That being said, good luck find a set of general prior rules and scaling them. [0] http://science.sciencemag.org/content/338/6111/1202 http://science.sciencemag.org/content/338/6111/1202 [1] https://medium.com/@seanaubin/why-does-ai-still-suck-and-can-it-suck-less-9db36be294dc#.akb82s8o9 https://medium.com/@seanaubin/why-does-ai-still-suck-and-can...