The Link Between Meditation and Artificial Intelligence
Sit down, put your feet on the ground, sit up straight, close your eyes, breath deeply, start to notice your thoughts. Now slowly separate your awareness from your thoughts, in other words take a couple steps back in your mind from your thoughts, notice the thoughts but don’t actually think them, start to listen and watch the thoughts as the arise.
What you are doing is becoming an observer of your thoughts, not participating in them, simply acknowledging them, letting them go along as you remain, a non-critical observer.
It may help if you had a word to repeat (like ‘ohm’, or any mantra) or a spot to concentrate your eyes on (with your eyes closed) to help you keep your observer self separate from your thoughts.
Soon, you may notice all sorts of thoughts, ideas, even images and sounds show up, from somewhere, and if you let them be (without judging or participating in them), fade away.
This practice of observing thoughts is fascinating and amusing. As the observer, I noticed how many of the thoughts I noticed seemed random and unrelated, some of them even seemed foreign to me.
The above is how I typically explain what it’s like to meditate.
Soon after I woke up this morning, right before I left the bed, I had an idea that connected the above experience of meditation with designing artificial intelligence software. The insight I gained from meditating, into how we are observers of seemingly random thoughts, seemed appropriately suited for artificial intelligence computer programs.
Let me clarify, I’m trying to say that artificial intelligence design may benefit (and become actualized) if and when we can mimic what human minds do. In other words, when random ideas are picked from an idea bank or generated in some fashion then presented to an observing software for split-second analysis. By generating or picking up thousands of seemingly random pieces of information, then running them through the examining software (observer) that can discard some ideas and pick others for further consideration, we may be able to give a machine the ability to find inspiration and creativity.
I am posting this idea online, in hopes that someone can pick it up and utilize it, I hope you can find this helpful, assuming that this concept is not already being utilized in AI projects.
My meditation practice allowed me to see how thoughts, almost at random, criss-crossed my mind, and how consciousness or the observer can pick some and discard others, unconsciously processing countless numbers of ideas behind the scene of the logical mind. This is very different than the typical linear and object oriented computer programming, it is also different than fuzzy logic in computer science.
We, humans, make sense and create order out of seeming chaos in our heads, with ease (unconsciously), maybe this is what AI programs have been lacking.
Let me hear your thoughts, leave a comment below. Thanks!!
Added Feb 18, 2010:
I mentioned traditional linear, object oriented programming and fuzzy logic above. I would like to clarify that fuzzy logic (ie. going with degrees of truth instead of true/false values) can be a valuable part of the AI strategy mentioned above. The observer program would utilize fuzzy logic to evaluate all the random ideas, images, concepts, sounds being presented to it. The source of such ideas, concepts etc being presented can also be another complex program, or multiple ones, I will call the source the ‘Presenter’ program(s); the observing program is what I’ve referred to as ‘Observer’.
What programming language used is not the topic of this article, rather the mechanism or flow of data, or the inner works of an AI brain.
All the ideas evaluated by the Observer (using fuzzy logic and other evaluation mechanisms) are stored in their original form, in their altered forms, and whatever evaluation process and outcome performed is also stored, whatever associations made to other ideas are also stored, only to be recycled again in the future by being used by the Presenter program and re-presented at random (and in logical order at times) to the Observer to be re-examined and re-processed. In this way ideas grow, get altered, and new concepts get formed. In other words, we got AI !
All the ideas evaluated by the Observer (using fuzzy logic and other evaluation mechanisms) are recycled, they are stored in their original form, in their altered forms, and whatever evaluation process and outcome performed is also stored, whatever associations made to other ideas are also stored, only to be recycled again in the future by being used by the Presenter program and re-presented at random (and in logical order at times) to the Observer to be re-examined and re-processed. In this way ideas grow, get altered, and new concepts get formed. In other words, we got AI !
I decided to refer to this mechanism as the EZDAPO mechanism. I wrote this article to put this idea out in the public domain, you are free to use it, as long as you refer to where you found this idea (i.e. give credit to the source of this idea, this blog)
Read my comments below this article for more details.
Did you enjoy this article?
I found your post very interesting! Hope you’ll continue to post. Much love.
I mentioned traditional linear, object oriented programming and fuzzy logic above. I would like to clarify that fuzzy logic (ie. going with degrees of truth instead of true/false values) can be a valuable part of the AI strategy mentioned above. The observer program would utilize fuzzy logic to evaluate all the random ideas, images, concepts, sounds being presented to it. The source of such ideas, concepts etc being presented can also be another complex program, or multiple ones, I will call the source the ‘Presenter’ program(s); the observing program is what I’ve referred to as ‘Observer’.
What programming language used is not the topic of this article, rather the mechanism or flow of data, or the inner works of an AI brain.
All the ideas evaluated by the Observer (using fuzzy logic and other evaluation mechanisms) are stored in their original form, in their altered forms, and whatever evaluation process and outcome performed is also stored, whatever associations made to other ideas are also stored, only to be recycled again in the future by being used by the Presenter program and re-presented at random (and in logical order at times) to the Observer to be re-examined and re-processed. In this way ideas grow, get altered, and new concepts get formed. In other words, we got AI !
mentioned traditional linear, object oriented programming and fuzzy logic above. I would like to clarify that fuzzy logic (ie. going with degrees of truth instead of true/false values) can be a valuable part of the AI strategy mentioned above. The observer program would utilize fuzzy logic to evaluate all the random ideas, images, concepts, sounds being presented to it. The source of such ideas, concepts etc being presented can also be another complex program, or multiple ones, I will call the source the ‘Presenter’ program(s); the observing program is what I’ve referred to as ‘Observer’.
What programming language used is not the topic of this article, rather the mechanism or flow of data, or the inner works of an AI brain.
All the ideas evaluated by the Observer (using fuzzy logic and other evaluation mechanisms) are stored in their original form, in their altered forms, and whatever evaluation process and outcome performed is also stored, whatever associations made to other ideas are also stored, only to be recycled again in the future by being used by the Presenter program and re-presented at random (and in logical order at times) to the Observer to be re-examined and re-processed. In this way ideas grow, get altered, and new concepts get formed. In other words, we got AI !
Another comment on the article above, here I give the mechanism of Presenter and Observer a title (EZDAPO) and I discuss how the EZDAPO mechanism can function by utilizing different types of logic (classical logic, and non-classical logic as in fuzzy logic or even futuristic emotional logic)
Comment:
When I think about the ideas in this article, it all is really simple. People don’t always analyze things as black/white or good/evil or true/false, especially people that think outside the box. Those of us that are free of dogma, usually (attempt to) examine things from different angles and sometimes know that truth is (and other labels are) relative and can change based on the observer, situation, time, and other factors.
This is where fuzzy logic, alongside classical logic can be used. In fact, it would be amazing if we could utilize ‘emotional logic’. Emotional logic would be more representative to actual human logic, most people’s unconscious minds function based on emotions and emotional logic not mathematical, or even fuzzy logic. The unconscious mind also functions based on beliefs, associations, habits and other ‘programming’.
As far as I know, we do not have ‘emotional logic’ in computing, yet. I believe emotional logic would require both fuzzy logic and classical logic, coupled with a new element (i.e. an emotional value).
The article above and the strategy I presented (the mechanics/mechanism of Presenter and Observer) will remain intact regardless of what logic is used (be it classical, fuzzy or futuristic emotional logic). This mechanism depends on behind the scenes ultra-quick processing of massive amounts of data presented by a Presenter program (this data being random at time, new — as in newly created at times–, meaningless at times, and related to other ideas at times)
I think it would be appropriate to call the mechanism I presented in the article EZDAPO. EZ stands for easy (and are also my initials), DA stands for data, P stands for Presenting, O stands for Observing. Put together the full title for this mechanism of simulating the human mind is: EZ Data Presenting-Observing Mechanism.
The engineering and programming behind the Observing software, will likely vary. I expect many different approaches at ‘Observing’ the presented data, not only using different types of logic and programming languages but also different approaches, concepts, workflows, ranging from simple to astronomically complex. However I’d prefer simple designs than can be replicated to create complex, almost organic systems (as in fractal design). Naturally, cloud based computing and organic computing can also utilize this EZDAPO system.
The EZDAPO concept itself is simple, I truly believe that the most ingenious ideas are the simple ones. In fact, simplicity is the key element in everything beautiful.
Naturally, we can compound (have multiples of) these simple EZDAPO mechanisms to compound the results. These EZDAPO mechanisms can interact with each other, or not, they can work in parallel or in sequence or in any number infinite possible arrangements. EZDAPO can be used to represent the unconscious mind, and the conscious mind. Results from one module can be put into another.
It is reasonable to speculate, if we had 2 identical EZDAPO mechanisms (or modules) starting with the same data set or data bank, that are left to process data for some time (using whatever kind of logic be it fuzzy, classical and emotional logic,) that the outcomes from these 2 EZDAPO mechanisms would definitely differ. This introduces uniqueness to this system; I believe uniqueness is a key element (or outcome) of any artificial intelligence (AI) system.
This idea came to me as a result of my experience of being an observer of my thoughts while meditating. I don’t know if the idea of the EZ Data Presenting-Observing Mechanism (EZDAPO) is an original one. If this is a new idea, I hope that someone out there would utilize it.
I wrote this to put the idea out in the public domain, you are free to use it, as long as you refer to where you found this idea (i.e. give credit to the source of this idea, this blog).