Search FQXi

Please also note that we do not accept unsolicited posts and we cannot review, or open new threads for, unsolicited articles or papers. Requests to review or post such materials will not be answered. If you have your own novel physics theory or model, which you would like to post for further discussion among then FQXi community, then please add them directly to the "Alternative Models of Reality" thread, or to the "Alternative Models of Cosmology" thread. Thank you.

Forum Home
Introduction

Order posts by:
chronological order
most recent first

Display:
all posts
member posts highlighted
member posts only

RECENT POSTS IN THIS TOPIC

Jennifer Nielsen: on 5/30/18 at 20:04pm UTC, wrote Essentially I think for a being to be so-called "conscious" or have...

Jennifer Nielsen: on 5/30/18 at 20:02pm UTC, wrote I think that to understand something is to be a part of something and to...

Steve Dufourny: on 11/26/16 at 11:07am UTC, wrote Helo to both of you, Mr Hosein, Siddartha Gottam Buddah d say that after...

Georgina Woodward: on 11/25/16 at 21:37pm UTC, wrote Hi Nicholas, the probabilities may be to do with measurement outcomes which...

Nicholas Hosein: on 11/22/16 at 14:57pm UTC, wrote Reality is contemplative. That is, it is the object of contemplation by...

John Prytz: on 11/30/14 at 13:10pm UTC, wrote THE RISE AND RISE OF THE MACHINE It would appear based on relatively...

Christophe Tournayre: on 4/25/14 at 11:23am UTC, wrote Dear Ian, Could understanding be a sense? Similar to eyesight or hearing? ...

Alice lewis: on 3/24/14 at 11:04am UTC, wrote I think there's something else entirely to it than basically Gödelian...

FQXi FORUM
September 26, 2021

CATEGORY: Ultimate Reality [back]
TOPIC: Understanding, artificial intelligence, and so forth [refresh]

Member Ian Durham wrote on Mar. 7, 2010 @ 02:13 GMT
In the discussion over the nature of mathematics, one interesting side-argument that recently started was over the idea of "understanding." What does it take to really understand something? The answer is not as simple as you'd think. In fact it is a hotly debated topic in artificial intelligence (AI) circles. J.R. Searle gave the following argument - called the Chinese room argument - against...

view entire post

report post as inappropriate
Christophe Tournayre replied on Apr. 25, 2014 @ 11:23 GMT
Dear Ian,

Could understanding be a sense? Similar to eyesight or hearing?

From an evolutionary perspective, we humans may have developed this ability to perceive nature/our environment through another way. At first, our understandings were weak, fuzzy, black or white, but as this sense developped, our ability to perceive our environment became richer, varied and even colorful though it is still a long way to the deepness of visual perception.

Understanding is contextual to the environment. By changing hypothesis (or input), one can understand different objects. It is similar to a man who looks at a city from different places.

This idea brings understanding to the real physical world. There is nothing magical, simply an external reality that we look at from where we stand.

Christophe

report post as inappropriate

Pankaj Seth wrote on Mar. 7, 2010 @ 14:04 GMT
Whatever "understanding" is, it cannot be achieved via thought, as there is always a further "why" to deal with… ad infinitum.

report post as inappropriate

Anonymous wrote on Mar. 7, 2010 @ 16:57 GMT
Isn't this simply a reforumalation of Icompletenes, in the spirit of Godel? To understad something, you need to form relationhips among the members and the rules which associte them. In the cause of language, you need defintions, which are essentialy axiomatic assignments that we use to create the mental imagery in our thoughts that corresponds to the objects we are experiencing through the senses. You can never simply take the relations we form among the members and use them to derive the existence and structre of the defintions themselves.

report post as inappropriate

Member Ian Durham wrote on Mar. 7, 2010 @ 18:05 GMT
Maybe, but I think there's more to it than simply Gödelian incompleteness. There is clearly a link to emergence and potentially to consciousness here and I fail to see how Gödel's theorems shed any light on that.

report post as inappropriate

Anonymous wrote on Mar. 7, 2010 @ 19:42 GMT
I dissagree. This is a tour-de-force example of the incompleteness of any axiomatic system to account for itself. Ultimately, the ecamples you presented are an isssue of association. Asking the person to understand Chinese from the relations given is the same as asking a computer to reproduce the fundmanetal axioms that it was supplied by using only those axioms.

report post as inappropriate

Pankaj Seth wrote on Mar. 8, 2010 @ 00:48 GMT
Lets start from the beginning. A human being feels the absolute mysteriousness of his own existence, and of the world in which he exists. Immanuel Kant said "I feel awe at two things, the starry skies above me and the moral universe within me.". Even though one's existential situation is not clear, one has, somehow, a moral sense allowing one to navigate the world populated with others. Somehow, there is the ability to recognize what one ought to do in this or that situation. What would it mean to understand one's situation? I suppose that it would mean to have some sort of narrative in one's mind about one's origin, how one is constituted and perhaps one's end. Even if one cannot obtain that in a way that is satisfying, it is anyway, not necessary in order to be a morally upright person. This is the context in which "understanding" is interesting to me. Many narratives have been created, and some of these have led to either the strengthening or the weakening of the moral sense, fostering either a helping or fearing/attacking of others.

So, we build a narrative about self and world, wanting it to be full of veracity, not contradicting any facts which can be established. We attempt to build a 'Scientia' of 'Physis', a knowledge of nature, in which we are embedded. What are the limits to this endeavour? How far can our narrative be sketched out? How far can thought go? Many, many facts can and have been discovered, but the big picture involves piecing these facts together into a grand narrative. Can the narrative ever be something other than Mythos, though filled with facts?

report post as inappropriate

James Putnam wrote on Mar. 12, 2010 @ 21:09 GMT
Dear Ian,

"...If it is true that a systems approach is the answer, at what point does a system become complex enough to exhibit true intelligence? This is an interesting question related to emergence. But where does emergence occur? ..."

"...It seems that there is an important connection between the notion of true "understanding" and emergence & complexity. ..."

I will reply to the idea of 'understanding' in a follow-up message. I am replying to the ideas of emergence and complexity with regard to how they relate to intelligence. I think that intelligence cannot 'emerge'. The concept of emergence, I think, is a substitute to avoid admitting that we do not know the fundamental properties from which intelligence is generated. By generated, I mean that those fundamental properties must pertain directly to intelligence itself or they can never lead us to an understanding of intelligence.

The notion of emergence merely represents the point at which we are forced to abandon our mechanical notions of the universe and admit that we and intelligence are the most important properties that this universe has given birth to,and, that we do not know how or why. It is the point where our theoretical ideas clearly fail us. I think that this point of being forced to abandon our mechanical ideas is synonymous with the probable fact that it means our mechanical ideas are artificial right from the start.

Complexity is a building process. However, it can only form structures that are already implied in the fundamental properties that that complexity is built upon. Theoretical physics has nothing to offer us with regard to the property of intelligence. Artificialo intelligence is just what its name implies. It is totally artificial. It is our pretense or false claim to have the ability to mechanically produce the property of intelligence. Any theoretical effort to imply that intelligence can emerge magically from fundamental properties that represent pure dumbness should be summarily rejected as a logical fallacy.

James

report post as inappropriate

Member Ian Durham wrote on Mar. 13, 2010 @ 03:49 GMT
James,

I think I have to disagree with you on the point that theoretical physics has nothing to offer us in relation to this.

First, I think it is important that we recognize that there is a difference between intelligence and consciousness. Where "understanding" comes into this, is not entirely clear. But, at any rate, it does seem to be the case that the more complex an organism is, the more intelligence intelligence it seems to possess. Note, however, that this seems to be restricted to a certain class of "organisms," i.e. there seems to be something special about biology. Until we understand what that is, we won't really be able to make much progress toward the "Holy Grail" of AI.

Ian

report post as inappropriate
Pankaj Seth replied on Mar. 13, 2010 @ 22:45 GMT
Ian,

Mind and matter are an inseparable whole (QM), so how can one exist without or before the other? Therefore, how can one emerge from the other? Mind and life are always discerned together; where there is life, there is mind also. So, what applies to the emergence of mind also applies to the emergence of life. Thus, how could life emerge from matter, if matter and mind are inseparable?

Are we at an intractable impasse?

report post as inappropriate

James Putnam replied on Mar. 29, 2010 @ 19:31 GMT
Ian,

I think I should move slowly, selecting one point at a time. You said:

"...Note, however, that this seems to be restricted to a certain class of "organisms," i.e. there seems to be something special about biology. ..."

I think the word 'restricted' is the most revealing part of this statement. I think it is an important part of the evidence that theoretical physics is not helpful and is instead detrimental to solving the problem of the existence of intelligence. In my opinion, it is theoretical physics that is the cause of the appearance of this restriction and separation of biological properties from the rest of the universe.

I think that empirical physics and any mathematical equations that properly model the patterns seen in empirical evidence is capable of helping us remove the problem of 'restricting' our understanding of the origin and evolution of intelligence.

I think the basic problem is that when theoretical physics is included, it brings with it an interpretation of the properties of the universe that are 'restricted' to dumbness. Neither intelligence nor awareness, and any other suggestion that the fundamental properties of the universe are anything other than dumb and purposeless without intelligent meaning, is permitted for consideration due only to the imposition of a philosophical preference that is introduced, not for scientific reasons, but for serving humanistic attitudes.

I do not say this to promote religion. I say this to enable us to see that the fundamentals of the universe are responsible for the intelligence necessary to make decisions such as prefering one philosophical preference over another. If we disregard human emotional needs and concentrate on the evolution of the universe, then I think we should recognize that all effects, including all intelligent effects, must have been provided for in potential form since the beginning of the universe.

James

report post as inappropriate

James Putnam replied on Mar. 30, 2010 @ 19:35 GMT
Ian,

Following on my previous message, I think that the universe does not have the logical breaks that we tend to see in its operation. I think that all such divisions are artificial and are of our own making. Life, intelligence, and the rest of the universe shared their origin.

"...Where "understanding" comes into this, is not entirely clear. ..."

I would say that 'understanding' is not clear at all. The first problem encountered when studying the development of understanding is the recognition that we receive a storm of tiny, multitudinous numbers of truncated almost random signs of events that have occurred. We call these tiny signs 'photons'. There is no way to find meaning in that storm unless we already know what to look for and already contain the means to understand it. The means for understanding must exist first. That 'means' had the same origin as did everything else.

Understanding in its macroscopic form is attached to complex forms of life. However, it is also attached to complex arrangements of particles of matter. It arises from the cooperative properties of that 'matter'. We have no problem seeing that all other types of results arise from the cooperative properties of matter. However, we do not know why any of these results arise. Theoretical physics offers us a way to view the mechanical type results. However, the existence of intelligent life demonstrates that matter cannot be what theoretical physics tells us it is.

James

report post as inappropriate

Pankaj Seth wrote on Mar. 13, 2010 @ 14:34 GMT
Analysis is diabolical; that is, it divides the whole into parts, and they remain unjoined until the the symbolical joining process is undertaken. Scientific knowledge has become so specialized that in different corners different persons work on very specific intellectual pursuits, while very, very few have enough breadth of knowledge to even attempt to rejoin the parts into a whole.

In one corner of science, there is the problem of understanding of how mind arises from matter (brain). In another corner, there is the problem of understanding how mind and matter form an inseparable whole (QM). Not for nothing did Bohr say "Those who look at QM and are not shocked have not understood it." Feynman said "This is the only mystery." This is a schizoid situation that we are in, and it is crazy making. Its not trivial to say that what overall picture science gives us determines our future. There are numerous other examples of how much we have broken the unity of self/nature. We are creating fragmented pictures of ourselves, other creatures and existence as a whole.

The Shamanic cultures and persons who daily experience the unity of self/nature are often branded "pre-scientific" as a pronouncement of the superiority of the scientific cultures. But from the outside, it is we who look fragmented, tormented and violent.

In one corner, people are trying to establish what linear progression in time has led to the present (Evolution). In another corner, other people have already established that time does not enjoy an ontological status, that it is a relative metric not to be understood as self-existent (Relativity). In yet another corner, space and time have been shown to not even have a noumenal status (QM).

In one corner of human intellectual activity, it is considered that survival is the overarching dynamic of existence, because… survival happens. In another corner, people have established that there is a hierarchy of dynamics with survival being only one of the lot… other dynamics elucidated include the establishing of relationships which even trump one's own survival; further, there is also expression, creativity and transcendent self-knowledge which have been established as dynamics every bit as important as survival and competition.

How shall we a construct a overall narrative that somehow encompasses all of this, and more? If we perform poorly in the Symbolical, joining endeavour, we will only deepen and extend our own suffering, as well as harming those nearby that we share the world with.

report post as inappropriate

Saibal Mitra wrote on Mar. 21, 2010 @ 01:20 GMT
Another issue that is i.m.o. far more interesting than the Chinese Room Argument is the following. If you have an intelligent machine, you can arrange it to be run in a fully determinstic environment. Eg., just simulate its brain inside a virtual environment.

Now, you can then argue that the physical states it evolves through can be mapped one to one to those of a simple system that runs...

view entire post

report post as inappropriate
Member Ian Durham replied on Mar. 26, 2010 @ 20:43 GMT
Now that's *really* interesting. I will have to think some more about it (I'm what is called a "processor" rather than a "reactor" I guess) but, at first glance, it seems like an idea worth thinking about in greater depth.

report post as inappropriate

Saibal Mitra replied on Apr. 5, 2010 @ 17:12 GMT
Observers as Hamiltonians can perhaps also be motivated directly from the MWI of quantum mechanics. It has been argued that the way the classical world effectively emerges is due to decoherence. Of course, this picture has to be true at a certain level, decoherence certainly exists and unless quantum mechanics as we know it is somehow wrong, I don't see what could be wrong with this picture.

But on a deeper level there seems to be something not quite right. If we consider the entire multiverse, then the Schrödinger equation for that can be argued to reduce to H|psi) = 0. Time can only really exist at a local level, at least that is the most natural picture considering relativity. But if the entire multiverse is a state |psi) which is an eigenvector of the full Hamiltonian, then the information about the excited states of H is not present in |psi) itself.

This means that the observations of local observers cannot be extracted from

|psi) itself. I think that this fact is recognized by the researchers in quantum gravity; they seem to invoke some extra structure in the multiverse to make individual universes with nontrivial dynamics well defined.

report post as inappropriate

Member Ian Durham replied on Apr. 6, 2010 @ 20:16 GMT
Interesting points. Note, however, that H|psi) = 0 (which is the Wheeler-DeWitt equation), can also be used to model closed time-like curves. I have a preprint floating around out there (which will hopefully get cleaned up and published at some point) that notes that it is possible to have temporal evolution with such a Hamiltonian. It merely needs to be cyclic in nature.

I don't think MWI is necessarily the answer, though, nor do I think decoherence is the only explanation. There's a statistical argument for it to: the more complex a system is, the less likely it is to be perfectly reversible (some of the details behind this idea go back to Eddington in the 1920s).

report post as inappropriate

Lev Goldfarb wrote on Jul. 23, 2010 @ 22:40 GMT
I would like to draw your attention to my paper "Representational Formalism in Which Syntax and Semantics Are Congruent: Towards the Resolution of Searle's Chinese Room Challenge".

By the way, I do believe that the issue raised there is relevant not just to AI but to physics too, i.e. for any science, the formalism in which syntax and semantics are congruent is much more preferable.

report post as inappropriate
Joe Fisher replied on Aug. 7, 2010 @ 15:00 GMT
Never mind Artificial Reality, let me waste a bit more of your time by pointing out the absurdity of the Real Reality. Why is it that our senses, which are the monitors of our individual reality, only act instantaneously? The only reality that exists is the reality of the here and the now. There and then could not have been created by any omnipotent God ever. There and then could never have been created out of nothing 13.7 billion years ago. There is no way that here and now could have evolved out of the there and then. Here and now always has to be dynamically active. Only here and now truly exists. Wherever you go and whatever you do, you will only be able to travel in the here and now, and you will only be able to do whatever it is that you are doing here and now. The creator of the Artificial Reality device will always know that it is an Artificial Reality device that he/she has created, because unlike real reality, the Artificial Reality will have to have the logical construction only obtainable from theories concerning the there and then.

report post as inappropriate

Rainsmith replied on Sep. 18, 2010 @ 22:28 GMT
Yiu need to distinguish between percieved reality and reality in general. Our perception is a sample of a broader process. Who knows what happens to present once we 'drive past it', It is reasonable to conclude that it plays a role in the foundations of the future, our consciousness perhaps a wave travelling across an ocean of awareness.

report post as inappropriate

Georgina Parry replied on Nov. 4, 2010 @ 10:01 GMT
Joe, all,

Our senses do not act instantaneously. We do not perceive here and now but always observe a seamlessly stitched mental image that is a "patchwork" of delayed information giving an experience of here and now. Various pasts derived from photon data of different delay brought together by the mind into a present experience of space- time.

The present is not a single...

view entire post

report post as inappropriate

Alice lewis wrote on Mar. 24, 2014 @ 11:04 GMT
I think there's something else entirely to it than basically Gödelian deficiency. There is obviously a connection to development and possibly to awareness here and I neglect to perceive how Gödel's hypotheses shed any light on that.

report post as inappropriate

John Prytz wrote on Nov. 30, 2014 @ 13:10 GMT
THE RISE AND RISE OF THE MACHINE

It would appear based on relatively recent trends that sooner rather than later Homo sapiens will morph into Homo robotus. Just as likely, Homo robotus will share the world or will be supplanted by Robot robotus. It’s all due to the rise and rise of the machine.

Where would we be without the intelligence to develop technology? Where will we be with...

view entire post

report post as inappropriate

Nicholas I Hosein wrote on Nov. 22, 2016 @ 14:57 GMT
Reality is contemplative. That is, it is the object of contemplation by brains existing within it (for whatever purpose). If it were not contemplative, science and Philosophy would be non-existent. It also remains intact and coherent, assuming a reality external to observation. Quantum Indeterminacy says that our calculations of the future may only be probabilistic. This means there is a barrier to how accurate and how much knowledge we can possess. What are the implications to the epistemic accuracy of our contemplations of reality?

report post as inappropriate
Georgina Woodward replied on Nov. 25, 2016 @ 21:37 GMT
Hi Nicholas, the probabilities may be to do with measurement outcomes which are not correlated to those states prior to measurement.There may not be a state for that 'property' prior to measurement. Thinking of the Stern Gerlach apparatus - the exposure of a particle to the apparatus and measurement protocol provokes a new (random) outcome, if the particle has not been exposed to the same test prior. The same test prior produces the same outcome as before. Each new orientation produces a random outcome each same test the same outcome.'Properties' are not retained over different tests. That seems to indicate the property isn't an intrinsic property but manifestation of an environmental/contextual influence.

report post as inappropriate

Steve Dufourny replied on Nov. 26, 2016 @ 11:07 GMT
Helo to both of you,

Mr Hosein, Siddartha Gottam Buddah d say that after all the contemplations and compassion are torchs of this universal infinite entropy.Like the universalims is the sister of altruism.That said unfortunately we loose our contemplations on this earth.....Globally speaking.The sufferings and the disasater and disorders are our realityactual and sad.It is very sad because the solutions exist towards the points of equilibrium.What a world dear thinkers what a world...But we evolve also ....Hopes......

report post as inappropriate

Jennifer L Nielsen wrote on May. 30, 2018 @ 20:02 GMT
I think that to understand something is to be a part of something and to have truly interacted with it.

I'm currently working on a model of AI, or really AA--artificial awareness--I'm basing on a principle I call "strong immersion". I believe in order for something to become truly aware, and thus gain true understanding, an object has to be truly immersed in its environment perceptually. I first began developing the idea around 2016 and believe the idea is somewhat of a topology problem.

report post as inappropriate
Jennifer L Nielsen replied on May. 30, 2018 @ 20:04 GMT
Essentially I think for a being to be so-called "conscious" or have apparent autonomy (what looks like "free will") it must be a global non-equilibrium sensory system folded in on itself and immersed in its environment to an extent that it processes the outside as its inside and the inside as its outside. To appear to have free will it must make its "decisions" based on global non-equilibrium responses to its outer environment (which is also topologically part of its inner environment). Programming based on digitally modelling exclusively pre-assigned choices will result in an automaton dependent on another global sensory non-equilibrium life form immersed in its environment. (Programming based on pre-signed choices represents a language, which can be executed by living systems and used to direct other beings and automaton objects into apparent cultures, but not to replicate sensory experience or integrate an automaton into a situation allowing apparent autonomous choice; language is not sensate and not conscious, but dependent upon sensate interpreters).

report post as inappropriate