Cortical Learning Algorithms with Predictive Coding for a Systems-Level Cognitive Architecture

David Rawlinson
Nov 9, 2014 · 1 min read

This is a quick post to link a poster paper by Ryan McCall, who has experimented with a Predictive-Coding / Cortical Learning Algorithm (PC-CLA) hybrid approach. We found the paper via Ryan writing to the NUPIC theory mailing list.

What’s great about the paper is it links to some of the PC papers we mentioned in a previous post and covers all the relevant literature, with clear and detailed descriptions of key features of each method.

So we have Lee & Mumford, Rao and Ballard, Friston (Generalized Filtering)… It’s also nice to see Baar’s Global Workspace Theory and LIDA (a model of consciousness or, at least, attention).

Ryan has added a PC-CLA module to LIDA and tested robustness to varying levels of input noise. So, early days with the experiments but great start.

Originally published at Project AGI.

Project AGI

Combining principles of biological cognition and machine learning to devise, empirically test and prove novel algorithms that can achieve general intelligence.

David Rawlinson

Written by

Project AGI

Combining principles of biological cognition and machine learning to devise, empirically test and prove novel algorithms that can achieve general intelligence.