Thanks it is very handy to get something that is compatible with SUMO.
JohnDavidBustard
Karma: 159
Thank you for the thoughtful comments. I am not certain that the approach that I am suggesting will be successful but I am hoping that more complex experiences may be explainable from simpler essences, similar to the behaviour of fluids from simpler atomic rules. I am currently focused on the assumption that the brain is similar to a modern reinforcement learning algorithm where there is a one or more large learnt structures and a relatively simple learning algorithm. The first thing I am hoping to look at is if all the concious experiences could be explained purely by behaviours associated with the learning algorithm. Even better if in trying to do this it indicates new structures that the learning algorithm should take. For example, we have strong memories of sad events and choices we regret, this implies we rank the importance of past experiences based on these situations and weight them more heavily when learning from them. We might avoid a stategy because our intuition says it makes us sad (it is like other situations that made us sad) rather than it simply being a poor strategy to achieve our goals.
Great links, thank you, I hadn’t considered the drug effects before that is an interesting perspective on positive sensations. Also I wanted to say I am a big fan of your work, particularly your media synthesis stuff. I use it in teaching of deep learning to show examples of how to use academic source code to explore cutting edge techniques.
Perfect, thank you
[Question] Any taxonomies of conscious experience?
A high level post on its use would be very interesting.
I think my main criticism of the Bayes approach is that it leads to the kind of work you are suggesting i.e. have a person construct a model and then have a machine calculate its parameters.
I think that much of what we value in intelligent people is their ability to form the model themselves. By focusing on parameter updating we aren’t developing the AI techniques necessary for intelligent behavior. In addition, because correct updating does not guarantee good performance (because the model properties dominate) then we will always have to judge methods based on experimental results.
Because we always come back to experimental results, whatever general AI strategy we develop its structure is more likely to be one that searches for new ways to learn (with bayesian model updating and SVMs as examples) and validates these strategies using experimental data (replicating the behaviour of the AI field as a whole).
I find it useful to think about how people solve problems and examine the huge gulf between specific learning techniques and these approaches. For example, to replicate a Bayesian AI researcher an AI needs to take a small amount of data, an incomplete informal model of the process that generates it (e.g. based on informal metaphors of physical processes the author is familiar with) and then find a way of formalising this informal model (so that its behaviour under all conditions can be calculated) and possibly doing some theorem proving to investigate properties of the model. They then apply potentially standard techniques to determine the models parameters and judge its worth based on experiment (potentially repeating the whole process if it doesn’t work).
By focusing on Bayesian approaches we aren’t developing techniques that can replicate these kinds of lateral and creative thinking behaviour. Saying there is only one valid form of inference is absurd because it doesn’t address these problems.
I feel that trying to force our problems to suit our tools is unlikely to make much progress. For example, unless we can model (and therefore largely solve) all of the problems we want an AI to address we can’t create a “Really Good Model”.
Rather than manually developing formalisations of specific forms of similarity we need an algorithm to learn different types of similarity and then construct the formalisation itself (or not as I don’t think we actually formalise our notions of similarity and yet can still solve problems).
Automated theorem proving is a good example where the problems are well defined yet unique, so any algorithm that can construct proofs needs to see meta patterns in other proofs and apply them. This brings home the difficulty of identifying what it means for things to be similar and also emphasises the incompleteness of a probabilistic approach: the proof that the AI is trying to construct has never been encountered before, in order for it to benefit from experience it needs to invent a type of similarity to map the current problem to the past.
Eh not impossible… just very improbable (in a given world) and certain across all worlds.
I would have thought the more conventional explanation is that the other versions are not actually you (just very like you). This sounds like the issue of only economists acting in the way that economists model people. I would suspect that only people who fixate on such matters would confuse a copy with themselves.
I suspect that people who are vulnerable to these ideas leading to suicide are in fact generally vulnerable to suicide. There are lots of better reasons to kill yourself that most people ignore. If you think you’re at risk of this I recommend you seek therapy, thought experiments should not have such drastic effects on your actions.
Thanks for your reference it is good to get down to some more specific examples.
Most AI techniques are model based by necessity: it is not possible to generalise from samples unless the sample is used to inform the shape of a model which then determines the properties of other samples. In effect, AI is model fitting. Bayesian techniques are one scheme for updating a model from data. I call them incomplete because they leave a lot of the intelligence in the hands of the user.
For example, in the thesis reference the author designs a model of transformations on handwritten letters that (thanks to the authors intelligence) is similar to the set of transformations applied to numeric characters. The primary reason why the technique is effective is because the author has constructed a good transformation. The only way to determine if this is true is through experimentation, I doubt the bayesian updating is contributing significantly to the results, if another scheme such as an SVM was chosen I would expect it to produce similar recognition results.
The point is that the legitimacy or otherwise of the model parameter updating scheme is relatively insignificant in comparison to the difficulty in selecting a good model in the first place. As far as I am aware, as there are a potentially infinite set of models, Bayesian techniques cannot be applied to select between them, leaving the real intelligence being provided by the user in the form of the model. In contrast, SVMs are an attempt to construct experimentally useful models from samples and so are much closer to being intelligent in the sense of being able to produce good results with limited human interaction. However, neither technique addresses the fundamental difficulty of replicating the intelligence used by the author in creating the transformation in the first place. Fixating on a particular approach to model updating when model selection is not addressed is to miss the point, it may be meaningful for gambling problems but for real AI challenges the difference it makes appears to be irrelevant to actual performance.
I would love to discuss what the real challenges of GAI are and explore ways of addressing them, but often the posts on LW seem to focus on seemingly obscure game theory or gambling based problems which don’t appear to be bringing us closer to a real solution. If the model selection problem can’t be addressed then there is no way to guarantee that whatever we want an AI to value, it won’t create an internal model that finds something similar (like paperclips) and decides to optimise for that instead.
Silently down voting criticism of Bayesian probability without justification is not helpful either.
From what I understand, in order to apply Bayesian approaches in practical situations it is necessary to make assumptions which have no formal justification, such as the distribution of priors or the local similarity of analogue measures (so that similar but not exact predictions can be informative). This changes the problem without necessarily solving it. In addition, it doesn’t address the issue of AI problems not based on repeated experience, e.g. automated theorem proving. The advantage of statistical approaches such as SVMs is that they produce practically beneficial results with limited parameters. With parameter search techniques they can achieve fully automated predictions that often have good experimental results. Regardless of whether Bayesianism is the law of inference, if such approaches cannot be applied automatically they are fundamentally incomplete and only as valid as the assumptions they are used with. If Bayesian approaches carry a fundamental advantage over these techniques why is this not reflected in their practical performance on real world AI problems such as face recognition?
Oh and bring on the down votes you theory loving zealots :)
Thank you very much for your great reply. I’ll look into all of the links. Your comments have really inspired me in my exploration of mathematics. They remind me of the aspect of academia I find most surprising. How it can so often be ideological, defensive and secretive whilst also supporting those who sincerely, openly and fearlessly pursue the truth.
Thank you, my main goal at the moment is to get a handle on statistical learning approaches and probability. I hope to read Jaynes’s book and the nature of statistical learning theory once I have some time to devote to them. however I would love to find an overview of mathematics. Particularly one which focuses on practical applications or problems. One of the other posts mentioned the Princeton companion to Mathematics and that sounds like a good start. I think what I would like is to read something that could explain why different fields of mathematics were important, and how I would concretely benefit from understanding them.
At the moment I have a general unease about my partial mathematical blindness, I understand the main mathematical ideas underlying the work in my own field (computer vision) and I’m pretty happy with the subjects in numerical recipes and some optimisation theory. I’m fairly sure that I don’t need to know more, but it bothers me that I don’t. At the same time I don’t want to spend a lot of time wading through proofs that are unlikely to ever be relevant to me. I have also yet to find a concrete example in AI where an engineering approach with some relatively simple applied maths has been substantially weaker than an approach that requires advanced mathematical techniques, making me suspect that mathematics is as it is because it appeals to those who like puzzles, rather than necessarily providing profound insight into a problem. Although I’d love to be proved wrong on that point.
So, assuming survival is important, a solution that maximises survival plus wireheading would seem to solve that problem. Of course it may well just delay the inevitable heat death ending but if we choose to make that important, then sure, we can optimise for survival as well. I’m not sure that gets around the issue that any solution we produce (with or without optimisation for survival) is merely an elaborate way of satisfying our desires (in this case including the desire to continue to exist) and thus all FAI solutions are a form of wireheading.
One frustration I find with mathematics is that it is rarely presented like other ideas. For example, few books seem to explain why something is being explained prior to the explanation. They don’t start with a problem, outline its solution provide the solution and then summarise this process at the end. They present one ‘interesting’ proof after another requiring a lot of faith and patience from the reader. Likewise they rarely include grounded examples within the proofs so that the underlying meaning of the terms can be maintained. It is as if the field is constructed so that it is in the form of puzzles rather than providing a sincere attempt to communicate idea as clearly as possible. Another analogy would be programming without the comments.
A book like Numerical Recipies, or possibly Jaynes book on probability, is the closest I’ve found so far. Has anyone encountered similar books?
I’m not sure I understand the distinction between an answer that we would want and a wireheading solution. Are not all solutions wireheading with an elaborate process to satisfy our status concerns. I.e. is there a real difference between a world that satisfies what we want and directly altering what we want? If the wire in question happens to be an elaborate social order rather than a direct connection why is that different? What possible goal could we want pursued other than the one which we want?
Ok, so how about this work around.
The current approach is to have a number of human intelligences continue to explore this problem until they enter a mental state C (for convinced they have the answer to FAI). The next stage is to implement it.
We have no other route to knowledge other than to use our internal sense of being convinced. I.e. no oracle to tell us if we are right or not.
So what if we formally define what this mental state C consists of and then construct a GAI which provably pursues only the objective of creating this state. The advantage being that we now have a means of judging our progress because we have a formally defined measurable criteria for success. (In fact this process is a valuable goal regardless of the use of AI but it now makes it possible to use AI techniques to solve it).
Interesting, if I understand correctly the idea is to find a theoretically correct basis for deciding on a course of action given existing knowledge and then to make this calculation efficient and then direct towards a formally defined objective.
As distinct from a system which potentially sub optimally, attempts solutions and tries to learn improved strategies. i.e. one in which the theoretical basis for decision making is ultimately discovered by the agent over time (e.g. as we have done with the development of probability theory). I think the perspective I’m advocating is to produce a system that is more like an advanced altruistic human (with a lot of evolutionary motivations removed) than a provably correct machine. Ideally such a system could itself propose solutions to the FAI problem that would be convincing, as a result of an increasingly sophisticated understanding of human reasoning and motivations.
I realise there is a fear that such a system could develop convincing yet manipulative solutions. However the output need only be more trustworthy than a human’s response to be legitimate (for example based on an analysis of its reasoning algorithm it appears to lack a Machiavellian capability, unlike humans).
Or put another way, can a robot Vladimir (Eliezer etc.) be made that solves the problem faster than their human counterparts do. And is there any reason to think this process is less safe (particularly when AI developments will continue regardless)?
If there is an answer to the problem of creating an FAI, it will result from a number of discussions and ideas that lead a set of people to agreeing that a particular course of action is a good one. By modelling psychology it will be possible to determine all the ways this can be done. The question then is why choose one over any of the others? As soon as one is chosen it will work and everyone will go along with it. How could we rate each one? (they would all be convincing by definition). Is it meaningful to compare them? Is the idea that there is some transcendent answer that is correct or important that doesn’t boil down to what is convincing to people?
When I say feel, I include:
I feel that is correct. I feel that is proved etc.
Regardless of the answer, it will ultimately involve our minds expressing a preference. We cannot escape our psychology. If our minds are deterministic computational machines within a universe without any objective value, all our goals are merely elaborate ways to make us feel content with our choices and a possibly inconsistent set of mental motivations. Attempting to model our psychology seems like the most efficient way to solve this problem. Is the idea that there is some other kind of answer? How would could it be shown to be legitimate?
I suspect that the desire for another answer is preventing practical progress in creating any meaningful solution. There are many problems and goals that would be relatively uncontroversial for an AI system to attempt to address. The outcome of the work need only be better than what we currently have to be useful we don’t have to solve all problems before addressing some of them and indeed without attempting to address some of them I doubt we will make significant progress on the rest.
Ok, I certainly agree that defining the goal is important. Although I think there is a definite need for a balance between investigation of the problem and attempts at its solution (as each feed into one another). Much as how academia currently functions. For example, any AI will need a model of human and social behaviour in order to make predictions. Solving how an AI might learn this would represent a huge step towards solving FAI and a huge step in understanding the problem of being friendly. I.e. whatever the solution is will involve some configuration of society that maintains and maximises some set of measurable properties from it.
If the system can predict how a person will feel in a given state it can solve for which utopia we will be most enthusiastic about. Eliezer’s posts seem to be exploring this problem manually, without really taking a stab at a solution, or proposing a route to reaching one. This can be very entertaining but I’m not sure it’s progress.
Thanks for the comment, I think it is very interesting to think about the minimum complexity algorithm that could plausibly be able to have each conscious experience. The fact that we remember events and talk about them and can describe how they are similar e.g. blue is cold and sad, implies that our internal mental representations and the connections we can make between them must be structured in a certain way. It is fascinating to think about what the simplest ‘feeling’ algorithm might be, and exciting to think that we may someday be able to create new conscious sensations by integrating our minds with new algorithms.