What do you mean by “prejudice”? The “textbook definition” basically amounts to “applying Bayesian priors to humans” and that doesn’t seem like a bad thing.
basically amounts to “applying Bayesian priors to humans
There is nothing about Bayes in the “textbook definition”. It boils down to “applying strong priors to humans” where “strong” means “resistant to change by evidence”.
The “textbook definition” basically amounts to “applying Bayesian priors to humans” and that doesn’t seem like a bad thing.
The OED says “Preconceived opinion not based on reason or actual experience; bias, partiality; (now) spec. unreasoned dislike, hostility, or antagonism towards, or discrimination against, a race, sex, or other class of people.” The further definitions given are either shades of this one or other senses not relevant here (e.g. legal terminology).
From a brief glance at the web, other dictionaries say the same. The second half of the OED’s definition is but a currently prominent instance of the first half. That part is probably what you mean by “the textbook definition”, but I don’t know what textbooks you’ve been reading. Probably books by progressives that you study to keep your wrath warm.
“Not based on reason or actual experience.” “Unreasoned.” That is the core of the concept, is it not?
In Bayesian reasoning, that, without the pejorative overtones, is what your prior is. Your state of belief, represented as a probability distribution, before you have seen the data to which you intend to apply Bayesian reasoning.
I am not seeing that in your use of the phrase “Bayesian prior”, which you seem to be waving as a rationalist password without noticing the step that it implies, of looking at data and updating from it. Without that, it is not a prior — there is nothing that it is prior to. No, for you “applying Bayesian priors to humans” means stopping at your priors without any awareness that a prior is an expression of ignorance to be improved on, not knowledge to be clung to.
That part is probably what you mean by “the textbook definition”, but I don’t know what textbooks you’ve been reading.
The definition I learned in public school, which does have a rather extreme “progressive” bias.
I am not seeing that in your use of the phrase “Bayesian prior”, which you seem to be waving as a rationalist password without noticing the step that it implies, of looking at data and updating from it.
Like the data on the relationship between sex and intelligence. The data on the relationship between how many men a women has had sex with and her ability to participate in future stable relationships.
Like the data on the relationship between sex and intelligence. The data on the relationship between how many men a women has had sex with and her ability to participate in future stable relationships.
In that case, you are talking about posteriors, not priors, and there is no need for the Bayes jargon. Beliefs, conclusions, from whatever sources and methods it may have been. “Bayes” is not a Power Word: Stun.
Of course, it’s still prior to looking at the person in front of you and observing them.
I tend to ascribe a naïve etymology of pre-judgement to ‘prejudice’, so I suppose that is the sense I was using it there, but I really wasn’t appealing to any “textbook definition” I know of.
What do you mean by “prejudice”? The “textbook definition” basically amounts to “applying Bayesian priors to humans” and that doesn’t seem like a bad thing.
There is nothing about Bayes in the “textbook definition”. It boils down to “applying strong priors to humans” where “strong” means “resistant to change by evidence”.
Ok, so what evidence was AA refusing to update on?
I’m not talking about AA, I’m talking about your understanding of prejudice.
The OED says “Preconceived opinion not based on reason or actual experience; bias, partiality; (now) spec. unreasoned dislike, hostility, or antagonism towards, or discrimination against, a race, sex, or other class of people.” The further definitions given are either shades of this one or other senses not relevant here (e.g. legal terminology).
From a brief glance at the web, other dictionaries say the same. The second half of the OED’s definition is but a currently prominent instance of the first half. That part is probably what you mean by “the textbook definition”, but I don’t know what textbooks you’ve been reading. Probably books by progressives that you study to keep your wrath warm.
“Not based on reason or actual experience.” “Unreasoned.” That is the core of the concept, is it not?
In Bayesian reasoning, that, without the pejorative overtones, is what your prior is. Your state of belief, represented as a probability distribution, before you have seen the data to which you intend to apply Bayesian reasoning.
I am not seeing that in your use of the phrase “Bayesian prior”, which you seem to be waving as a rationalist password without noticing the step that it implies, of looking at data and updating from it. Without that, it is not a prior — there is nothing that it is prior to. No, for you “applying Bayesian priors to humans” means stopping at your priors without any awareness that a prior is an expression of ignorance to be improved on, not knowledge to be clung to.
The definition I learned in public school, which does have a rather extreme “progressive” bias.
Like the data on the relationship between sex and intelligence. The data on the relationship between how many men a women has had sex with and her ability to participate in future stable relationships.
In that case, you are talking about posteriors, not priors, and there is no need for the Bayes jargon. Beliefs, conclusions, from whatever sources and methods it may have been. “Bayes” is not a Power Word: Stun.
Of course, it’s still prior to looking at the person in front of you and observing them.
It is, however, often used to fill in the phase 2 in the underpants gnomes business plan.
Good, I see you are making progress in understanding this.
I hope that one day I will be able to say the same of you.
Unfortunately, I cannot say the same of you.
I tend to ascribe a naïve etymology of pre-judgement to ‘prejudice’, so I suppose that is the sense I was using it there, but I really wasn’t appealing to any “textbook definition” I know of.
For people you haven’t interacted with it isn’t, for other people it’s the posteriors you should apply, not the priors.