A utility-maximizing varient of AIXI

Thanks for putting in all this work!

I recommend you post this to the MAGIC list and ask for feedback. You'll also see a discussion of Anja's post there.

I like how you specify utility directly over programs, it describes very neatly how someone who sat down and wrote a utility function

$U\(\\.{y}\\.{x}\_{<k}y\underline{x}\_{k:m\_k}\$ )

would do it: First determine how the observation could have been computed by the environment and then evaluate that situation. This is a special case of the framework I wrote down in the cited article; you can always set

$U\(\\.{y}\\.{x}\_{<k}y\underline{x}\_{k:m\_k}\$ =\sum_{q:q(y_{1:m_k})=x_{1:m_k}}%20U(q,y_{1:m_k}))

This solves wireheading only if we can specify which environments contain wireheaded (non-dualistic) agents, delusion boxes, etc..

[-]AlexMennen13y10

True, the U(program, action sequence) framework can be implemented within the U(action/observation sequence) framework, although you forgot to multiply by 2^-l(q) when describing how. I also don't really like the finite look-ahead (until m_k) method, since it is dynamically inconsistent.

This solves wireheading only if we can specify which environments contain wireheaded (non-dualistic) agents, delusion boxes, etc..

Not sure what you mean by that.

[-]Anja13y20

you forgot to multiply by 2^-l(q)

I think then you would count that twice, wouldn't you? Because my original formula already contains the Solomonoff probability...

[-]AlexMennen13y10

Oh right. But you still want the probability weighting to be inside the sum, so you would actually need $U\(\\\.\{y\}\\\.\{x\}\_\{<k\}y\\underline\{x\}\_\{k:m\_k\}\$ =\frac{1}{\xi\left(\dot{y}\dot{x}_{%3Ck}y\underline{x}_{k:m_{k}}\right)}\sum_{q:q(y_{1:m_k})=x_{1:m_k}}%20U(q,y_{1:m_k})2%5E{-\ell\left(q\right)}%0A)

[-]Anja13y20

True. :)

[-]Anja13y00

Let's stick with delusion boxes for now, because assuming that we can read off from the environment whether the agent has wireheaded breaks dualism. So even if we specify utility directly over environments, we still need to master the task of specifying which action/environment combinations contain delusion boxes to evaluate them correctly. It is still the same problem, just phrased differently.

[-]AlexMennen13y00

If I understand you correctly, that sounds like a fairly straightforward problem for AIXI to solve. Some programs q_1 will mimic some other program q_2's communication with the agent while doing something else in the background, but AIXI considers the possibilities of both q_1 and q_2.

[-]V_V12y40

Formulas aren't being rendered correctly, they appear as latex code.

[-]AlexMennen12y00

Looks like it's back up now.

[-]AlexMennen12y00

That appears to be because the formulas are images hosted by codecogs.com, which is down.

[-]Alex_Altair13y00

This is great!

I really like your use of $U\(q,y\_\{1:\\infty\}\$ ). The seems to be an important step along the way to eliminating the horizon problem. I recently read in Orseau and Ring's "Space-Time Embedded Intelligence" that in another paper, "Provably Bounded-Optimal Agents" by Russell and Subramanian, they define $V\(\\pi, q\$ %20=%20u(h(\pi,%20q))) where $h\(\\pi, q\$ ) generates the interaction history $ao\_\{1:\\infty\}$ . (I have yet to read this paper.)

Your counterexample of supremum chasing is really great; it breaks my model of how utility maximization is supposed to go. I'm honestly not sure whether one should chase the path of U = 0 or not. This is clouded by the fact that the probabilistic nature of things will probably push you off that path eventually.

The dilemma reminds me a lot of exploring versus exploiting. Sometimes it seems to me that the rational thing for a utility maximizer to do, almost independent of the utility function, would be to just maximize the resources it controlled, until it found some "end" or limit, and then spend all its resources creating whatever it wanted in the first place. In the framework above we've specified that there is no "end" time, and AIXI is dualist, so there are no worries of it getting destroyed.

There's something else that is strange to me. If we are considering infinite interaction histories, then we're looking at the entire binary tree at once. But this tree has uncountably infinite paths! Almost all of the (infinite) paths are incomputable sequences. This means that any computable AI couldn't even consider traversing them. And it also seems to have interesting things to say about the utility function. Does it only need to be defined over computable sequences? What if we have utility over incomptuable sequences? These could be defined by second-order logic statements, but remain incomputable. It gives me lots of questions.

[-]Cole Wyeth9mo10

The draft I recently sent you answers some of these questions. The expected utility should be an appropriately defined integral not a sum.

[-]AlexMennen13y00

I'm honestly not sure whether one should chase the path of U = 0 or not. This is clouded by the fact that the probabilistic nature of things will probably push you off that path eventually.

Making the assumption that there is a small probability that you will deviate from your current plan on each future move, and that these probabilities add up to a near guarantee that you will eventually, has a more complicated effect on your planning than just justifying chasing the supremum.

For instance, consider this modification to the toy example I gave earlier. Y:={a,b,c}, and if the first b comes before the first c, then the resulting utility is 1 - 1/n, where n is the index of the first b (all previous elements being a), as before. But we'll change it so that the utility of outputting an infinite stream of a is 1. If there is a c in your action sequence and it comes before the first b, then the utility you get is -1000. In this situation, supremum-chasing works just fine if you completely trust your future self: you output a every time, and get a utility of 1, the best you could possibly do. But if you think that there is a small risk that you could end up outputting c at some point, then eventually it will be worthwhile to output b, since the gain you could get from continuing to output a gets arbitrarily small compared to the loss from accidentally outputting c.

There's something else that is strange to me. If we are considering infinite interaction histories, then we're looking at the entire binary tree at once. But this tree has uncountably infinite paths! Almost all of the (infinite) paths are incomputable sequences. This means that any computable AI couldn't even consider traversing them. And it also seems to have interesting things to say about the utility function. Does it only need to be defined over computable sequences? What if we have utility over incomptuable sequences? These could be defined by second-order logic statements, but remain incomputable. It gives me lots of questions.

I don't really have answers to these questions. One thing you could do is replace the set of all policies (P) with the set of all computable policies, so that the agent would never output an uncomputable action sequence [Edit: oops, not true. You could consider only computable policies, but then end up at an uncomputable policy anyway by chasing the supremum].

[-]Cole Wyeth9mo10

Whether the agent outputs an uncomputable sequence of actions isn't really a concern, but the environment frequently does (some l.s.c. environments output uniformly random percepts).

[-]Alex_Altair13y00

Making the assumption that...

Yeah, I was intentionally vague with "the probabilistic nature of things". I am also thinking about how any AI will have logical uncertainty, uncertainty about the precision of its observations, et cetera, so that as it considers further points in the future, its distribution becomes flatter. And having a non-dualist framework would introduce uncertainty about the agent's self, its utility function, its memory, ...

[-]Anja13y00

I think there is something off with the formulas that use policies: If you already choose the policy

$p:p\(x\_\{<k\}\$ =y_{%3Ck}y_k)

then you cannot choose an y_k in the argmax.

Also for the Solomonoff prior you must sum over all programs

$q:q\(y\_\{1:m\_k\}\$ =x_{1:m_k}) .

Could you maybe expand on the proof of Lemma 1 a little bit? I am not sure I get what you mean yet.

[-]AlexMennen13y00

If you already choose the policy ... then you cannot choose an y_k in the argmax.

The argmax comes before choosing a policy. In $\{\\displaystyle\\arg\\max\_\{y\_\{k\}\\in Y\}\\sup\_\{p\\in P:p\\left \\dot\{x\}\_\{k\}\\right=\\dot\{y\}\_\{k\}y\_\{k\}\}\.\.\.\}$ , there is already a value for y_k before you consider all the policies such that p(x_<k) = y_<k y_k.

Also for the Solomonoff prior you must sum over all programs

Didn't I do that?

Could you maybe expand on the proof of Lemma 1 a little bit?

Look at any finite observation sequence. There exists some action you could output in response to that sequence that would allow you to get arbitrarily close to the supremum expected utility with suitable responses to the other finite observation sequences (for instance, you could get within 1/2 of the supremum). Now look at another finite observation sequence. There exists some action you could output in response to that, without changing your response to the previous finite observation sequence, such that you can get arbitrarily close to the supremum (within 1/4). Look at a third finite observation sequence. There exists some action you could output in response to that, without changing your responses to the previous 2, that would allow you to get within 1/8 of the supremum. And keep going in some fashion that will eventually consider every finite observation sequence. At each step n, you will be able to specify a policy that gets you within 2^-n of the supremum, and these policies converge to the policy that the agent actually implements.

I hope that helps. If you still don't know what I mean, could you describe where you're stuck?