Here is an article written for you! What is Bayesianism?
My personal struggle is where this differs from ‘clear-headedness.’ I think that much of this website is geared towards helping us get closer to the ideal Bayesian, though the connections are not mentioned specifically.
Can anyone give an example of where they explicitly used Bayesian reasoning? It makes sense that it is right, but … unlike other things on this website that can be transferred into skills or habits. My guess is that having a deeper understanding of Bayesian probability would help with understanding what evidence is and how much confidence should be placed in what.
A separate confusion of mine is that in Eliezer’s explanation of Bayes theorem----I was able to do the math problems correctly and so I didn’t make whatever the usual mistake was. Because of this, I have knowledge of the right way to solve probability problems (at least if I spend a long time thinking about them), buI never went down the wrong path got slapped by having an Incorrect Answer. That doesn’t mean I won’t notice a mistake, but I think that learning things the wrong way helps you understand why they are wrong later. So my confusion is that I am never very confident as to whether I am doing things the “Bayesian way” or not. I’ve found that the Law of Conservation of Expected Evidence has been the most helpful in understanding the consequences of Bayesian reasoning, beyond solving math problems.
My awareness of Bayesian reasoning doesn’t quite enable me to use it explicitly with success most of the time, or maybe the successes are not vivid and spectacular enough to be noticed, but it does make me aware of Bayes-stupid inferences committed by me and others.
Just yesterday my father proclaimed that a certain beggar who tends to frequent our street with a kid or two and claim to be a homeless is a liar, because, well, he’s not a homeless because he is also often seen in a company of drunkards and he probably drags around the kids for show and they aren’t even his. I asked my dad whether the beggar’s claim of homelessness makes him more or less likely to be homeless. He said less likely, but after that he denied that the beggar’s failure to claim so would make him more likely to be a homeless.
I’m not sure I understand—why would he deny that the beggar’s failure to claim so would make him less likely to be homeless? I have trouble imagining how the conversation you’re describing went.
I personally feel like a deeper understanding of Bayesian probability has mainly just helped me to formalize things that are already obvious (the goal being to replicate what is obvious to humans in a computer, e.g. computer vision, robotics, AI, etc.). There have been few instances where it has actually helped me weigh evidence more effectively. But maybe I am missing some set of practical techniques.
Also, I was unable to parse the final paragraph that you wrote, would you mind re-stating it?
Here is an article written for you! What is Bayesianism? My personal struggle is where this differs from ‘clear-headedness.’ I think that much of this website is geared towards helping us get closer to the ideal Bayesian, though the connections are not mentioned specifically.
Can anyone give an example of where they explicitly used Bayesian reasoning? It makes sense that it is right, but … unlike other things on this website that can be transferred into skills or habits. My guess is that having a deeper understanding of Bayesian probability would help with understanding what evidence is and how much confidence should be placed in what.
A separate confusion of mine is that in Eliezer’s explanation of Bayes theorem----I was able to do the math problems correctly and so I didn’t make whatever the usual mistake was. Because of this, I have knowledge of the right way to solve probability problems (at least if I spend a long time thinking about them), buI never went down the wrong path got slapped by having an Incorrect Answer. That doesn’t mean I won’t notice a mistake, but I think that learning things the wrong way helps you understand why they are wrong later. So my confusion is that I am never very confident as to whether I am doing things the “Bayesian way” or not. I’ve found that the Law of Conservation of Expected Evidence has been the most helpful in understanding the consequences of Bayesian reasoning, beyond solving math problems.
Edited for clarity.
My awareness of Bayesian reasoning doesn’t quite enable me to use it explicitly with success most of the time, or maybe the successes are not vivid and spectacular enough to be noticed, but it does make me aware of Bayes-stupid inferences committed by me and others.
Just yesterday my father proclaimed that a certain beggar who tends to frequent our street with a kid or two and claim to be a homeless is a liar, because, well, he’s not a homeless because he is also often seen in a company of drunkards and he probably drags around the kids for show and they aren’t even his. I asked my dad whether the beggar’s claim of homelessness makes him more or less likely to be homeless. He said less likely, but after that he denied that the beggar’s failure to claim so would make him more likely to be a homeless.
I’m not sure I understand—why would he deny that the beggar’s failure to claim so would make him less likely to be homeless? I have trouble imagining how the conversation you’re describing went.
Uh, I mixed up a less likely and a more likely. Corrected.
In that case:
… the first bit should probably be “He said less likely”, in which case what you say makes much more sense.
I personally feel like a deeper understanding of Bayesian probability has mainly just helped me to formalize things that are already obvious (the goal being to replicate what is obvious to humans in a computer, e.g. computer vision, robotics, AI, etc.). There have been few instances where it has actually helped me weigh evidence more effectively. But maybe I am missing some set of practical techniques.
Also, I was unable to parse the final paragraph that you wrote, would you mind re-stating it?