Systems Neuroscience Approach to General Intelligence (SynAGI) - overview
SynAGI at IBM Research is a group of Neuroscientists, AI Researchers, and Computer Scientists with the common goal of designing artificial systems based on what we know about animal and human brains — and what we know about deep-learning-based artificial neural nets — then building novel AI architectures that exhibit behavior people would recognize as "general intelligence".
One of SynAGI's core goals is to design a deep learning system that implements Bernard Baars' model of the brain. Just as people for thousands of years believed engineered human flight would one day be possible because of birds, Norbert Weiner and others believed machine-based thought would be feasible because of brains. The feasibility of artificial thinking machines was elusive then, just as Daedalus never actually created a flying machine. Of course, science and engineering progressed to the era of the Wright brothers, and today, we believe AI technology and Neuroscience have progressed such that it’s again prudent to look to the brain as a model for AI. By examining the broader picture and overlap of current artificial neural networks, theoretical computer science, and systems neuroscience, we think we can make a plausible architecture and test it for certain clear cases where general intelligence is required. Even if success in our broadest agenda is elusive, we believe we can advance the state of current AI systems with this approach.
In the other pages linked here and undergoing construction, you'll find a more detailed description of our proposed solution, details about the problems it solves, descriptions of our testing strategy, and the observations from Neuroscience that support our assertion that we're on the right track.
We begin with a page that describes the literature that a researcher working with us might want to be familiar with.