Interested in math, Game Theory, etc.
Pattern
Sometimes companies end up selling a tool they originally developed for internal use. This post suggests that’s effective because it includes step 1 - creating something that generates value (for someone) - and makes it easier.
[Linkpost] The importance of stupidity in scientific research
I continue to be happy that few are reckoning with the implications of exponential growth, despite the resulting large quantity of obvious nonsense, because the reactions I would otherwise expect would cause far more harm than good.
I continue to be highly frustrated by the refusal to take testing or treatments seriously, especially given this (in my opinion wise) decision to mostly give up on containing Omicron. There’s always time to make things somewhat better, but mostly we have already missed our window.
Pretty sure you can’t have your cake and eat it too.
It’s a shame we can’t see the disagree number and the agree number, instead of their sum.
I enjoyed the MtG post by the way. It was brief, and well illustrated. I haven’t seen other posts that talked about that many AI things on that level before. (On organizing approaches, as opposed to just focusing on one thing and all its details.)
Open & Welcome Thread December 2021
I noticed the post had a “here is a thing/technique I find useful” component and a “here is an example of something like me using the thing/technique”. Did you disagree with the meta part (people talking about where they come from) or the specific part (where the author comes from)?
The bottom line is, adding “contrition” to TFT makes it quite a bit better, and allows it to keep pace with Pavlov in exploiting TFT’s, while doing better than Pavlov at exploiting Defectors.
This is no longer true if we add noise in the perception of good or bad standing; contrite strategies, like TFT, can get stuck defecting against each other if they erroneously perceive bad standing.
So cTFT moves TFT’s weakness to noise somewhere else. Where can we find real robustness?
From page 2 of the paper:
“cTFT is not the only evolutionarily stable rule which is Pareto-optimal (and hence yields the maximal pay-off if the whole population adopts it).”
We discuss cTFT, PAVLOV and REMORSE with analytical methods and numerical simulations, embedding them in a large class of stochastic strategies. Finally, we show that by replacing the conventions concerning the ‘‘standing’’ by another set (which is even easier to implement, and only depends on an ‘‘internal variable’’) one is led to a PRUDENT PAVLOV strategy which is an ESS and immune against errors both in implementing and in perceiving moves.
That sounds very useful for a population to have.
The problem in general, if you’re fond of strategies that “have short memories” but keep track of similar statistics instead:
page 11, being careful about bias:
In principle, one could apply other rules of ‘‘standing’’. To start with, we should replace this term by a more neutral one, in order not to get trapped by its connotations, and think only of an arbitrary ‘‘tagging’’ of the states without specifying which is ‘‘good’’ or ‘‘bad’’. A strategy is now specified by the probability to cooperate and/or change the standing in the next round, depending on the current state (including the current standing) of both opponents. It is plausible that we can obtain some evolutionarily stable strategies for many such codes.
12, after pPavlov’s implementation is explained.
It seems highly plausible that there exists a wide variety of workable ‘‘taggings’’ which yield interesting ESS’s. The question is whether an evolution based on mutation and selection would tend to lead to one form of ‘‘tagging’’ rather than another. This could ultimately shed light on why humans developed a sense of fairness, feelings of guilt, and highly effective social norms [see also Sugden (1986) and Young (1993) on the evolution of conventions]. The sheer combinatorial complexity of encompassing all conceivable codes, or taggings, is enormous, and the costs (in fitness) for reckoning with these ‘‘tags’’ seem difficult to evaluate. But it is a tempting problem.
Duplicates—digital copies as opposed to genetic clones—might not require new training (unless a whole/partial restart/retraining was being done).
When combined with self-modification, there could be ‘evolution’ without ‘deaths’ of ‘individuals’ - just continual ship of Theseus processes. (Perhaps stuff like merging as well, which is more complicated.)
I think this has been the longest and most informationally dense/complicated part of the sequence so far. It’s a lot to take in, and definitely worth reading a couple times. That said, this is a great sequence, and I look forward to the next installment.
This seriously disincentivizes ‘risky’ comments by accounts that have a good reputation. This can easily result in strategic-voting-like suboptimal outcomes.
Highlighting this bit. I hadn’t thought about this at all.
(I am not part of ‘the team’ btw.)
A few days ago, I came across a political screed posted here. If the rules are not merely, ‘no politics’, something stated more clearly, in an FAQ somewhere seems preferable. I’m sure this has been discussed a bit, but having some more clear guidelines that people can be directed to that are up to date (the old ‘no politics’ doesn’t seem clear enough) would be an improvement.
would you consider an chimpanzee to have moral agency? A gorilla? An orangutan? A gibbon? An elephant? A dolphin?
I believe the OP was asking if you do.
Theorist: I have a circuit.
Critic: Your circuit is broken.
Critic’s critic: But what will we do without calculators?
Maybe they think, all this AI stuff is just tools?
Maybe they’re more worried about other things. (Right now it’s easy to say Covid, the economy, stuff like that. Compare how seriously global warming is taken.)
Maybe it’s not someone’s job.
I mean, isn’t the answer to that, as laid out in the Sequences, that Rationality really doesn’t have anything to offer them?
I disagree. Here’s an example from the same piece:
people have a odd tendency to be okay with letting single random outcomes decide their success, even when it’s unnecessary.
I suspect if this is common in gaming it’s common in real life too. That people are getting so invested into singular outcomes because they’ve staked too much on them.
This is 1) testable and 2) actionable. Are there people who don’t need this advice? Perhaps. But could a lot of people use this? I think so. (The first time through I read this as “don’t just have one plan—it could fail. What will you do if it doesn’t work?”, though it’s more general than that.)
what use do I have for your ‘rationality’? Why should I change any of my actions from the societal default?”
I think letting someone else decide what your victory looks like (to you) is a really bad idea.
World’s First Octopus Farm—Linkpost to a Linkpost
Throw in being able to go home if they start sneezing, or feel like it.
I read the earlier posts and the conclusion section. It was great, and added some new things:
There’s also an interesting dynamic which is the 1->3 direct transition, while preserving the object level truth values. I think this is real and important, and seems to be missing from most maps of this space. I have definitely experienced dynamics, both in games and in real life, where people are primarily attempting to form coalitions and cause actions that are favorable to them, and choosing what to say mostly on that basis, but with the common knowledge expectation that lying is out of bounds and the object level still exists and positive sum actions are possible. There’s the version of this where everyone pays lip service to lying being out of bounds and lies their ass off anyway, occasionally punishing those who are blameworthy because they no longer have plausible deniability that they were lying. This isn’t that. This is the good kind of politics, where we have conflicting interests and also shared interests and we fight but also work together and stay friends after, because otherwise we get eaten by a lion.
Only physical books, no ebooks?