Open thread, September 25 - October 1, 2017

25th Sep 2017

10turchin

2[anonymous]

0morganism

0[anonymous]

2Thomas

2Oscar_Cunningham

1Lumifer

0g_pepper

0Lumifer

0g_pepper

0Lumifer

0g_pepper

0Osho

0abcdef

0Elo

0Khoth_duplicate0.7536578752539433

0Lumifer

0WalterL

0ChristianKl

0Lumifer

2gjm

2Lumifer

0abcdef

3IlyaShpitser

0abcdef

0username2

0IlyaShpitser

0MrMind

0abcdef

0MrMind

New Comment

30 comments, sorted by Click to highlight new comments since: Today at 11:14 PM

One coming this approaching spring will. This one was livestreamed but not sure if it was recorded.

An update to this was presented:

If you fail to get your n flips in a row, your expected number of flips on that attempt is the sum from i = 1 to n of i*2^-i, divided by (1-2^-n). This gives (2-(n+2)/2^n)/(1-2^-n). Let E be the expected number of flips needed in total. Then:

E = (2^-n)n + (1-2^-n)[(2-(n+2)/2^n)/(1-2^-n) + E]

Hence (2^-n)E = (2^-n)n + 2 - (n+2)/2^n, so E = n + 2^(n+1) - (n+2) = 2^(n+1) - 2

Up LW's alley: Physicists find we’re not living in a computer simulation and the link to the actual paper.

Interesting paper. But, contrary to the popular summary in the first link, it really only shows that simulations of certain quantum phenomena are impossible using *classical* computers (specifically, using the Quantum Monte Carlo method). But this is not really surprising - one area where quantum computers show much promise is in simulating quantum systems that are too difficult to simulate classically.

So, if the authors are right, we might still be living in a computer simulation, but it would have to be one running on a quantum computer.

Yep. This could be because Nick Bostrom's original simulation argument focuses on *ancestor* simulations, which pretty much implies that the simulating and simulated worlds are similar. However here, in question 11, Bostrom explains why he focused on ancestor simulations and states that the argument could be generalized to include simulations of worlds that are very different from the simulating world.

Is anyone interested in starting a small team (2-3 people) to work on this Kaggle dataset?

https://www.kaggle.com/c/porto-seguro-safe-driver-prediction

This is a pretty daunting takedown of the whole concept of political campaigning. It is pretty hilarious when you consider how much money, how much human toil, has been squandered in this manner.

From the link:

It’s an especially shocking result given the authors’ previous work. Kalla and Broockman conducted a large-scale canvassing experiment, published in 2016, that found that pro-trans-rights canvassers could change Miami residents' minds about transgender issues by having intense, substantive, 10-minute conversations with them. The persuasive effects of this canvassing were durable, lasting at least three months. ...

But now, Kalla and Broockman are finding that this kind of persuasion doesn’t appear to happen during campaigns, at least not very often.

I'd wait a couple of years, they'll probably change their mind again.

Besides, the goal of campaigning is not to change someone's mind -- it is to win elections.

On the face of it, the goal of campaigning is to win elections *by changing people's minds*.

It may also help e.g. by encouraging The Base, but if it turns out that that's the main way it's effective then I bet there are more effective means to that goal than campaigning.

Incidentally, if anyone's having the same nagging feeling I did -- weren't Kalla and Broockman involved somehow in some sort of scandal where someone reported on an intense-canvassing experiment like that but it was all faked, or something? -- the answer is that they were "involved" but on the right side: they helped to expose *someone else's* dodgy study, at the same time as they were doing their own which so far as I know is not under any sort of suspicion.

On the face of it, the goal of campaigning is to win elections by changing people's minds.

That doesn't look obvious to me unless we're talking not about the face but the facade. Campaigning is mostly about telling people what they want to hear, certainly not about informing them they will need to rearrange their prejudices [1].

From the elections point of view there are three groups of people you're concerned with:

Your own Rabid Base. You want to energise them, provide incentives for them to be loud, active, confident, with contagious enthusiasm.

Other parties' Rabid Bases. Flip the sign: you want to demoralise them, make them doubtful, weak, passive. You want them to sit inside and mope.

The Undecideds, aka the Great Middle through which you have to muddle. This is where most of the action is. Do you want to convince them with carefully arranged chains of logical policy arguments? Hell, no. They don't vote on this basis. They vote on the basis of (1) Who promises more; (2) Who seems to be less likely to screw the pooch; and (3) Who exhudes more charisma/leadership -- not necessarily in this order, of course. Most of this is System 1 stuff, aka the gut feeling.

Notice how pretty much none of the above involves *changing people's minds*.

[1] "A great many people think they are thinking when they are merely rearranging their prejudices" -- William James

I'm not a statistician, but I happen to have some intuitions and sometimes work out formulas or find them on the web.

I have a bunch of students that took a test each day. The test of each day had a threshold score out of, say, 100 points. Scores under the threshold are considered insufficient.

I don't know whether of the two is true:

I can either use the tests to evaluate the students, or the students to evaluate the tests.

I can evaluate the students using the tests and the tests using the students at the same time.

The option 2. seems counterintuitive at first sight, especially if one wants to be epistemically sound. It seems more intuitive at second sight, though. I think it might be analogous to how you can evaluate a circular flow of feedback by using linear algebra (cfr. LW 2.0 discussions).

Some other context: In my evaluation model I would rather not only consider whether the scores were sufficient or not, but consider how much they were sufficient or insufficient, possibly after opportunely transforming them. Also, I want the weights of the scores to decay exponentially. I would also rather use a bayesian approach.

Is this reasonable, and where can I find instructions on how to do so?

You have an experimental design problem: https://en.wikipedia.org/wiki/Design_of_experiments.

The way that formalism would think about your problem is you have two "treatments" (type of test, that you can vary, and type of student), and an "outcome" (how a given student does on a given test, typically some sort of histogram that's hopefully shaped like a bell).

Your goal is to efficiently vary "treatment" values to learn as much as possible about the causal relationship between how you structure a test, and student quality, and the outcome.

There's reading you can do on this problem, it's a classical problem in statistics. Both Jerzy Neyman and Ronald Fisher wrote a lot about this, the latter has a famous book.

In fact, in some sense this is *the* problem of statistics, in the sense that modern statistics could be said to have grown out of, and generalized from, this problem.

i do statistical consulting as part of my day job responsibilities, i'm afraid to say this is not how it works.

if you came to me with this question i would roll back to ask what exactly you are trying to achieve with the analyses, before getting into the additional constraints you want to include. unfortunately it's far more challenging if the data owner comes to the statistician after the data are collected rather than before (when principles of experimental design as ilya mentioned can be considered to achieve ability to successfully answer those questions using statistical methods).

that said, temporarily ignoring the additional constraints you mentioned (e.g. whether and how to transform data; exponential decay and what that actually means with respect to student evaluation scores; magic word "bayes") perhaps a useful search term would be "item response theory".

good luck

From a Bayesian perspective, you calculate P(S|T) and P(T|S) at the same time, so it doesn't really matter. What *does* matter, and greatly, are your starting assumptions and models: if you have only one for each entity, you won't be able to calculate how much some datum is evidence of your model or not.

Well, to calculate P(T|S) = p you need a model of how a student 'works', in such a way that the test's result T happens for the kind of students S with probability p. Or you can calculate P(S|T), thereby having a model of how a test 'works' by producing the kind of student S with probability p.

If you have only one of those, these are the only things you can calculate.

If on the other hand you have one or more complementary models (complemenetary here means that they exclude each other and form a complete set), then you can calculate the probabilities P(T1|S1), P(T1|S2), P(T2|S1) and P(T2|S2). With these numbers, via Bayes, you have both P(T|S) and P(S|T), so it's up to you to decide if you're analyzing stundents or tests.

Usually one is more natural than the other, but it's up to you, since they're models anyway.

## If it's worth saying, but not worth its own post, then it goes here.

Notes for future OT posters:

1. Please add the 'open_thread' tag.

2. Check if there is an active Open Thread before posting a new one. (

Immediatelybefore; refresh the list-of-threads page before posting.)3. Open Threads should start on Monday, and end on Sunday.

4. Unflag the two options "Notify me of new top-level comments on this article" and "