floobynewb

I have been following this for a while and have taken the time to understand it in a little detail. While it does not provide the 'answers', it does frame the question of 'what is an intelligent machine?' in a very precise manner. It's interesting work, what it needs is for someone to now work out how to build much better models and plug them into the framework provided by AIXI.

lol

Yes! Not enough people get this. Anonymous is a name for a loose, evolving affiliation of ideas, it is not a specific group of people. It personifies a set of beliefs, a view of the world, it allows a hive mind to express itself as an individual. It allows the ideas to speak for themselves. In so doing it allows those ideas to evolve more rapidly.

I am not well versed in history, so I can't say if this is novel, but it is a fabulous idea.

It is clear to me that, just as thought can emerge from the movement of charge between networks of neurons so can it emerge from the chatter of a million people. The same processes are at work, you might call it 'emergence' but I suspect that our mathematics does not yet capture it's description adequately.

This is the kind of system we need to develop and enhance if we are to create a better world. Our social structure is prescriptive and too rigidly hierarchical, it has broken away from it's dynamic, organic roots and lost touch with the magic that seems to generate flexible and resilient structure out of nothing.

If you accept the isomorphism between the mind and society, then you may see that the internet is radically disruptive. It has made communication orders of magnitude faster and it has changed the topology of the network described by society. This is changing us, quickly. For better or worse remains to be seen. But I suspect the effects of the internet revolution are only now beginning.

Perhaps I'm just seeing what I want to see...

The battle outside ragin' Will soon shake your windows And rattle your walls For the times they are a-changin'

(attribution should not be needed)

I honestly don't think you can make it much simpler. That is not to say it is simple, just that it is by nature not simple. You can dress the concepts up, make them look nice with a graphical front end, but really the complexity comes from the abstract notions implicit in functions to be performed. You can tailor things by building libraries of functions to perform common tasks, but these abstractions are never going to be sufficient. What is required is a method of translating from the world we are most familiar with, described by our natural language, into the world described by the language of the computer. The translator needs to understand the context of the natural language requirement and find its correct representation in binary. This is the job of the programmer. An automated solution may well be able to pass the Turing test. That being said, I often think about how to leverage machine learning to help bridge the language gap. Evolving an application is an interesting idea, where the requirements are drip fed to the system through the language of the UI the program should present. Translations from some physical representation, cogs, roads, queues, etc might be useful since it is instinctive to many people yet more precise in its descriptive power than spoken language.

I agree. It would certainly take decades for them to master the technology. It might even take them 50 years. A lot of the knowledge required is implicit in the industry taken as a whole. Put me in the stone age with a manual for making bronze. I would be proud of myself If I managed to forge something useful inside of a lifetimes work. Where does the ore come from? How do I extract it? Precisely how do I get the fire hot enough? Similarly, if I knew how to create a warp drive, there is absolutely no guarantee that I could actually build one with current day tech. It might take a decade just to produce one of the required components.

To be fair, I'm pretty sure he had his tongue firmly in his cheek. But point well made. Whenever this point comes up I think of a scene from 'the cube', is it a massive conspiracy or utter incompetence combined with some kind emergent process?

I suppose the question is: Are there appropriate abstractions which would bring conceptual simplicity to these systems? I find it hard to believe that there is not. A system which does not have such a property is more difficult for natural selection to operate upon. The changes which are likely to happen to a lineage over the course of time are likely to have evolved to be likely to move the lineage closer to a locally optimal phenotype. I do not think that a system which is incompressible, would display such dynamics, the system would be chaotic and small changes in the genotype would lead to divergent phenotypes wrt the fitness landscape. Perhaps the appropriate abstractions are spread out both temporaly and spatially.

The question of time is deeper than this. Why is it that your brain or for that matter any physical system is dependent on it's previous configurations? Why is it that there is such a thing as previous? The mathematical descriptions of reality we have devised are symmetrical in time...

I can imagine a universe where alternate number like systems are more appropriate as heuristics for day to day living. Why should every universe be easily described through isomorphisms to the infinite cyclic group? Why could a universe not be more easily understood (compressed) within the minds of it's inhabitants with systems which are non-abelian groups? Perhaps you could say that there can exist no universes where intelligent beings can exist which contain aspects which are compressible using such systems. But you would have to prove it.

I think the arrow of time question is very interesting, I would love to see and understand an answer to this before I die.

I experience this. But my addiction is for information, technical articles, scientific papers in mathematics, physics, machine learning, biology. Now I can do it all on my phone anywhere, anytime. Texting and social media are of little interest to me... Do you think this is still something I should worry about?

I was thinking along the same lines. What he is asking for is quite close to general AI!

I think this is wrong. Take for instance the task of classifying images. You can train a RBN with backprop (after contrastive divergence alg) to correctly classify images. In the process it has automagically determined properties of the image which allow it to perform the classification. These properties are combinations of pixel elements. So it has in effect determined how to solve a problem without your input. In a similar way, solving a set of simultaneous equations using any of a huge array of mechanical mathematical techniques is also solving a problem which you do not personally know how to solve. you could even consider using a Library of code as solving a problem you do not know how to solve...

Despite its disjointed nature I found that some interesting notions and feelings made their way out of the text and into my brain. I might even say that it was good. Definitely worth the effort for me.

Bang on.

I loved the overall site. I'm talking specifically about the text at the very bottom of the page, the small print section.

Ugh, read the small print section. Horrible, designed by committee to attempt to sound like their audience. I sometimes find these quite revealing, although irritating. It shows how they perceive the people they are targeting with a product. In this case we are flighty, feckless, caught up in a culture full of meaningless catch phrases. They have mistaken the irreverence and playfulness of the y,z gen for low brow incoherence.

Where you say schizophrenia you probably mean split personality disorder. I am not even bothering to re-read your first sentence because of your perpetuation of this horrendous misuse of language.

There really isn't. You're letting your bias influence your opinion. Read the paper, draw your conclusions carefully, I wager you will have to change them.

Damn you beat me to it.

This has occured to me too. Is there some extra dimension to all this that we cannot see? However I dont think its necessary to explain events.

I'm not sure I follow. There's nothing mystical about the idea. The hypothesis is testable as shown by the research. The data currently looks to support it. I'm not sure why you are saying it is unscientific. I grant you there may be alternative explanations for the data, but it is often true that science has been lead astray by incorrect framing of the question, but it's still science...

This is a real thing. There appear to be intimate connections between the mind and the body, in both directions. Levels higher up the hierarchy of dynamical systems within an organism appear to direct the function of lower systems. It's feedback loops all the way down.

I agree with this. To get out of this local trap we are going to need something revolutionary. This is not something you can plow money into, it will come, if indeed it ever comes, from left field. My bet is there is a new 'frame' to be found somewhere out in the land of mathematical abstraction. I think to solve this one we are going to have to get right down to the nitty gritty, where does complexity come from, how specifically does structure emerge from non structure? How can we design such systems?

My feeling on this:

There is no effective science for this. Determining a more effective structure, for such a horribly non-linear dynamical system, requires mathematics which we do not have. I suspect that we are operating within a conceptual framework which is poorly serving our purposes. I think that a good number of problems that we currently face are rooted in this poor framing. How do you effectively parallelise code? How do you create systems that effectively satisfy objectives within dynamic environments? How do you build general AI? What is the mechanism that underlies the function of the brain? Where does life come from, what is it's driver? What is computation? What precisely is complexity? And... how do you construct a society which is, in the ways that we prescribe, optimal?

I would wager that there is a single big fish out there that recasts these problems into a framework which renders them at least partially tractable. We are in limbo waiting for a genius to find us a way out of this mess.

I acutely remember this issue when I was maybe 10 - 14. I tried to talk to my parents about the sadness I was feeling in these areas, but I couldn't make them understand. I had no-one around me who understood what I was saying, no-one who could guide me through. It was a very difficult time and it took me into my late teens to really put these issues to bed.

I think many people have a hunch in this direction. But that may just be because we desire ourselves to be in some way special. I would be wary of anyone discounting it out of hand. What you are really asking is how does the brain work? If you really want to think sensibly about this you are going to need to:

1) be prepared to spend decades thinking about it. 2) become really, really good at maths. 3) get really, really good at physics. 4) get really, really good at computational neuroscience

and if you actually want an answer, you have two options:

1) wait for a genius to work it out 2) become the genius who works it out

Hacker news is not going to be able to help you on this one!

thank you

Deep link? The two fields both contain the notion of locality, big woop. The application of quantum nonlocal correlations to game theory is interesting, but this is not what is advertised in the article. I am displeased.

"Every nation gets the government it deserves." - Joseph de Maistre. Thank you, this quote suddenly really clicked for me. I had interpreted it as an indictment of apathy. It's actually a statement of fact, the government refelects society.

@gojomo, this is what I had thought the original poster meant. Shame it was a better, nicer joke that way around.

I actually find this plausible. Like the author I also have problems believing that random mutation and selection is a complete/adequate explanation. Just a hunch, there's a deeper theory to be had here.