What is Connectomic AI?

What is Connectomic AI? 
I define Connectomic AI as an Artificial Intelligent system based on connectomic/synaptic modeling. This can be created using actual connectomic maps or using the essence of connectomics. Unfortunately, complete connectomic maps are very scarce. C elegans has been around since 1986 (White et al) and there has been a great surge in mapping the neural connections in recent years. If you are aware of any connectomic data sets, complete or partial, please share in the comments. In another post, I will put together a list of sites. 

Practical Connectomic AI 
I believe by emulating connectomics, we can realize Artificial General Intelligence (AGI). I emulated the C elegans connectome in 2013-2014 and to my surprise, the robot acted like the biological worm.


The C elegans nervous system is a very simple 302 neurons. This emulation has been replicated many times using different robots and different programming languages. It is remarkable that just the connectome, or how the neurites are connected to one another, showed behaviors similar to how the worm moves through its environment. 

Let me breakdown it down +

The sensors on the robot were used to activate certain sensory neurons based on what the robot sensor represented. For example, the sonar on the robot, when within a certain distance from an obstacle, would begin stimulating sensory neurons that are associated to nose touch. This in turn would activate interneurons and eventually motor neurons. Since I did not have chemical sensors on the robot, I used the sound sensor where when it reached a certain decibel level, I would activate sensory neurons that were associated to the presence of food (thus the whistling in the video). 

The motor neuron activation to the 95 body muscles (motion) were collected and accumulated by whether it stimulated a Left or Right body muscle. The accumulated values were sent to the wheel motors, left/right, which gave the robot motion. 

In the video, you can see the robot approach the wall, stop, backup and turn to a different angle. There was no programming to tell the robot to do any of these behaviors. It was purely the connectome.

Another important observation is that the robot never did these steps exactly the same. It behaved like a biological animal. 

Working with the C elegans connectome and others, I have built robots that would avoid obstacles as they moved forward. Again, no direct programming to tell the robot there was an obstacle blocking the movement. It was just how the artificial nervous system that was wired that dictated the observed behaviors.

As I continue this blog, I will add much more about this initial experiment as well as my continued research, using both mapped animal connectomes and artificial connectomes.

Thanks for being here.

Comments

Post a Comment

Popular posts from this blog

Use Cases for the emulation of the Drosophila nervous system

 I am often asked, especially by VC types, what can you do with animal emulations? What they are really asking is how can this be applied and how can we make money? I will admit, emulating a nervous system is like telling people you are developing a new computer operating system. There are thousands of things you can run on an operating system so asking for the One thing, is analogous to asking what is the one thing you can run on your new OS that stands apart from all others? There is no One thing but thousands of things.  If I emulated a complete human level system, what is the One thing it can do that would exceed all others? The human brain is capable of thousands and thousands of things it can do. I think you get the picture.  The One thing that I have seen with all animal nervous system emulations is generalized intelligence. The one thing that current AI cannot accomplish which is being able to quickly adapt to any given situation and as we climb the evolutionary scale, we see
Read more

Making Dronesophila Part 1

Image
       E mulating the Drosophila nervous system was step one which I did successfully on a single laptop using functional groups. This works VERY well.   The functional groups I created were: Neuron Group    Synapse Count central                          1455116 optic_lobes                1217922 visual_projection      268547 AN                                    115292 Kenyon_Cell                92875 ALPN                          88302 CX                                    85831 medulla_intrinsic      81621 visual_centrifugal      76532 DN                                    58402 L1-5                          53746 ALLN                          34872 MBIN                          26501 bilateral                          25724 mechanosensory      19246 MBON                          18529 olfactory                          17689 LHCENT                12565 LHLN                          11217 gustatory                          7950 visual     
Read more

Creating Emulations of Nervous Systems

Image
    L et me begin this post by stating that what is contained within is a simplified idea of the actual coding and there are specifics I am leaving out due to the amount of information would be better contained in a book rather than a blog post. Here I am attempting to highlight concepts and not discuss in detail. Maybe later discussions will allow me to do so?  The key with emulations of nervous systems is real time emulation if you are wanting real life AGI The first order of emulation is machine constraint. When I first did connectomic experiments, pretty much every machine I worked with could only support 250 processes running at a given time (16-bit). I had lunch with a colleague one day in 2013 complaining about this limitation when he asked, "Don't you have a 32 bit system now?". I did so he told me that he bet I could have many more processes running under that architecture. When I got home, I created a quick app that started up 302 notepad.exe and there were no
Read more