Ternary and Artificial Intelligence

[jshep89]

So, I was wondering what everyone's thought are pertaining the application of ternary in the field of AI research? How it could be used to create an artificially intelligent machine/program.

I'm trying to replicate certain parts of the human brain onto a machine. For example the two most vital areas of the brain needed to be replicated onto a machine in order to achieve an artificially intelligent machine are the frontal lobe and temporal lobe.

In order to mimic the human minds decision making process a machine must have areas on its hardware to operate or perform functions that these areas of the brain do. However, this to me seems impractical to do on a binary based machine. Because it can only see data in two states. Which is why I believe ternary to be a more appropriate approach to mimic these functions.

What are your thoughts?

[Mac Buster]

As far as I know members of Computers lab in Moscow State University did the same thing several years ago. I remember corresponding report on our conference about results of the research. It seems they had success by using ternary DSSP to simulate proper decision making. As usually the experiment was applied to geometry theorem solving and produced better result than classic use of binary logic approaches held before.

[Shaos]

The only approach that can "mimic" brain functions is neural net that uses analog levels, so ternary is not applicable here...

But if we will use a higher level of logic the ternary approach could be helpful

[jshep89]

Mbr can you provide any links to that research?

[jshep89]

I suppose "mimic" was a bad choice of words. What I meant was to develop a way for machines to accomplish long term potentiation. Not in a way that is an exact copy of the process the human brain uses, but one that could be used for machines to create plans and make decisions.

With neural networks or at least the ones described in the link posted they are able to make decisions on the input and output right? However, it is still the user or the human providing the input that controls the machine correct? Or are they capable of obtaining information on their own?

Also, I didn't mean mimic the methods each part of the human brain uses to perform its function. I meant merely to mimic the function its self. Machines can already store memory and retrieve it, but there is nothing really there to decide what to do with that information (at least without human interaction.)

[Mac Buster]

Some words about AI:

- http://numenta.com/htmcla.php

[jshep89]

That is nothing short of amazing, but can ternary be applied to HTM? Or should it be applied?

[jshep89]

Can HTM be considered genetic programming or is it something else entirely?

Also, Shaos what do you mean a higher form of logic? Do you have something in mind?

[Shaos]

Also, Shaos what do you mean a higher form of logic? Do you have something in mind?

I meant something like Prolog or Lisp kind of thing...

[Shaos]

Google AI assistant Bard knows about Ternary computers

> Can you explain ternary computers?

Ternary computers are computers that use ternary logic instead of the more common binary logic. Ternary logic uses three values: 0, 1, and -1. This allows ternary computers to perform some operations more efficiently than binary computers.

Ternary computers have been around for many years, but they have never been widely adopted. There are a number of reasons for this. One reason is that ternary computers are more complex to design and build than binary computers. Another reason is that ternary computers are not as compatible with existing software and hardware as binary computers.

However, there are a number of potential advantages to using ternary computers. One advantage is that ternary computers can perform some operations more efficiently than binary computers. Another advantage is that ternary computers can be more energy-efficient than binary computers.

As the cost of designing and building ternary computers decreases, it is possible that we will see more widespread adoption of this technology. Ternary computers could be used in a variety of applications, including artificial intelligence, machine learning, and cryptography.

[Shaos]

Researchers upend AI status quo by eliminating matrix multiplication in LLMs
Running AI models without floating point matrix math could mean far less power consumption.
Benj Edwards - 6/25/2024, 3:27 PM
https://arstechnica.com/information-technology/2024/06/researchers-upend-ai-status-quo-by-eliminating-matrix-multiplication-in-llms/

The researchers' approach involves two main innovations: first, they created a custom LLM and constrained it to use only ternary values (-1, 0, 1) instead of traditional floating-point numbers, which allows for simpler computations. Second, the researchers redesigned the computationally expensive self-attention mechanism in traditional language models with a simpler, more efficient unit (that they called a MatMul-free Linear Gated Recurrent Unit—or MLGRU) that processes words sequentially using basic arithmetic operations instead of matrix multiplications.

Third, they adapted a Gated Linear Unit (GLU)—a gating mechanism to control information flow in neural networks—to use ternary weights for channel mixing. Channel mixing refers to the process of combining and transforming different aspects or features of the data the AI is working with, similar to how a DJ might mix different audio channels to create a cohesive song.

[Joeddc]

Ternary computing, which is based on three states (unlike the binary system that uses just two), is an interesting concept that could have potential applications in the field of Artificial Intelligence (AI). By enabling more efficient data representation and processing, ternary systems might offer advantages in terms of computational speed and power consumption. This could lead to advancements in AI, particularly in areas requiring massive data processing or optimization. The future of AI may involve exploring these alternative computing systems to create more powerful and energy-efficient technologies.

For instance, AI's integration into education is already showing its potential in improving personalized learning and offering intelligent tutoring systems. These advancements are transforming the educational landscape by making learning more accessible and tailored to individual needs. You can learn more about the impact of AI in education here: https://www.cleveroad.com/blog/ai-in-education-or-what-advantages-of-artificial-intelligence-in-education-you-can-gain-/

[Shaos]

This article doesn't exist:

Code: Select all

<Error>
<Code>NotFound</Code>
<Message>The resource you requested does not exist</Message>
<RequestId>98WF9KR2V816H6J1</RequestId>
<HostId>
kO5aei17Xh0nmMXba6p/5G3lDB5hDpDOcN1UZE6hf9L2fg0MEjmUgLdtxmJ0UArwPpCRGJlsHoMMfL1GFFPUOw==
</HostId>
</Error>