Mecha General Intelligence

 What is Mecha General Intelligence (MGI)? The easiest answer is the ability to navigate through any environment or a system that is domain agnostic. As I have stated often, the reason we have a nervous system is to be able to move. This has been echoed by many others who have also understood how motor function is the primary function of animals.

Behavior Science 101 teaches us that all animals have four basic functions, also called the 4 Fs, which are Feeding, Fleeing, Fighting and Mating. To perform anyone of the 4 Fs, an animal must move and must move no matter what the environmental conditions. We find these basic functions in the worm like C. elgans all the way up the evolutionary scale to homo sapiens. Having emulated lower animal nervous systems and applied to robotics, I have seen firsthand that these behaviors are displayed within a robot that is truly emulating an animal’s nervous system. I have used Connectomics/Synaptomics to perform these emulations with great success.

What is missing from current Artificial Intelligence (AI) and machine learning approaches is the element of curiosity. Curiosity is the strong desire to know or learn something. Curiosity is an emergent property of the 4 Fs. If an animal wants to feed, it must find food. This drive to find food can be summed up as a curiosity to discover where there is plenty of food. To find plenty of food, an animal must potentially move through a myriad of environmental factors. To move through domains of various conditions, an animal must have the ability to generalize in that domain. Nature has provided that ability through the animal nervous system which can be emulated and applied to robotics and/or application processes.

This generalization of movement applies to Fighting in that I must fight to maintain my abilities which could be in the form of direct threats or something like fighting through an environmental condition in order to satisfy a curiosity.  MGI applies to Fleeing. If I must move to avoid a threat, I better be able to quickly move through whatever obstacles are present. Of course, to Mate, no other gender is going to have much to do with me If I were to just lie there.

Current AI is a “monkey see, monkey do” technology (no offense to monkeys). There are no elements of curiosity within the technology and even an approach to try and simulate some form of curiosity to force the technology to try and learn on its own, will never have the continued drive animals have in order to survive and thrive. Creating a process that pushes a system to learn is not the same as an inherent process that continually advances the entity into always looking for better outcomes.

Therefore, Mecha General Intelligence is the foundation of Artificial General Intelligence (AGI) and without these behavioral attributes built into the design of the intelligence paradigm, AGI will always be a forced process and never truly have general intelligence.

There are some approaches other than my own animal nervous system emulations that could move to general intelligence. Bayesian Inference is one possibility and Verses.ai is pioneering this approach. I am not convinced the Bayesian Inference is the perfect match to instill curiosity into a learning system? The approach is based on reduction and MGI is not just about reduction; it is also about exploration (curiosity). Bayesian Inference may just be a better methodology to reduce errors and a more symbolic AI approach but may lack the means to create truly intelligence agents and machines. TBD.

MGI is a form of NeuroAI and I find that it is always fascinating that so many are trying  to find a means to simulate natural processes when Mother Nature has already given us a map. I believe that the emulation of animal nervous systems will give us the ability to create artificial nervous systems based on nature’s rules and apply these artificial nervous systems to robotics and other applications that will rival our intellectual abilities.  

MGI directly gives rise to AGI. Like nature, we just need to apply up an evolutionary scale.

 

Comments

Post a Comment

Popular posts from this blog

Building a Holographic Digit Classifier with NumPy and MNIST

Image
 In this post, we’ll explore how to build a simple, yet powerful, holographic memory system to classify handwritten digits from the popular MNIST dataset . Using NumPy , we’ll create a system that represents digits as holographic encodings —dense vector representations—and compares new images using cosine similarity . This approach draws inspiration from vector symbolic architectures (VSAs) and holographic reduced representations (HRRs) used in associative memory and cognitive modeling. 📦 Step 1: Organizing the Dataset Assume you have MNIST digit data where each digit ( 0 to 9 ) is stored in its own folder. Each image has been preprocessed and saved as a .npy file (a NumPy array), possibly as a vector encoding or 2D array. Folder structure: python-repl Copy Edit /mnist_vectors/ ├── 0/ │ ├── img_0.npy │ └── ... ├── 1/ │ └── ... ... ├── 9/ 🧪 Step 2: Create the Holographic Memory File We’ll create a single file ( digit_holograms.npz ) that stores a prototype ve...
Read more

What LLMs say about me: Timothy Busbice

 This was a pleasant surprise and I believe nailed me perfectly as to how I fit into AGI research. I encourage everyone to google themselves and do things like this to find out how the world views you. Prompt: where does Timothy Busbice fit in the AGI research world Timothy Busbice: A Proponent of Biologically Inspired Artificial General Intelligence In the diverse and dynamic world of Artificial General Intelligence (AGI) research, Timothy Busbice positions himself as a visionary advocate for a distinct, neuroscience-inspired approach. As the Chief Technology Officer of PROME, he champions "Biologic Intelligence," a methodology centered on the emulation of biological nervous systems, a field known as connectomics. This places him in a unique niche within the broader AGI landscape, which is currently dominated by large-scale deep learning and large language models (LLMs). Busbice's core thesis is that true general intelligence can be achieved by reverse-engineering ...
Read more

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 ...
Read more