We don't have a hundred biases, we have the wrong model • Hacker News Board

Re: We don't have a hundred biases, we have the wrong model

#21

Post by brrrrrm » Thu, Jul 21, 2022, 8:41 PM UTC

JoshCole wrote: post #16

If you apply the cognitive biases model to algorithms which have superhuman performance in various games - like AlphaZero, DeepBlue, Pluribus, and so on - the natural result is to conclude that these models are predictably irrational. The reason you get this conclusion is because it turns out to be necessary to trade off theoretical optimal answers for the sake of speed. The behavioral economic view of human irration…

> apply the cognitive biases model to algorithms which have superhuman performance in various games

Could you give an example of this?

Re: We don't have a hundred biases, we have the wrong model

#22

Post by toomim » Thu, Jul 21, 2022, 8:42 PM UTC

snapcaster wrote: post #15

toomim wrote: post #13

I wrote a PhD dissertation that made this point in 2013, and proposed a new "helocentric" economic model. The key shift is to move the utility function from evaluating a future state of the world to evaluating the utility of an opportunity for attention in the present moment . All the "cognitive errors" that we humans make are with respect to predicting the future. But we all know what we find appealing in the presen…

Thanks for sharing this. Very interesting idea

Thank you for saying so!

Re: We don't have a hundred biases, we have the wrong model

#23

Post by jjk166 » Thu, Jul 21, 2022, 8:47 PM UTC

The rational actor model assumes that a person will behave optimally - using all information available to make and carry out the best decision possible for their goals.

I strongly suspect that a better model is that people instead of optimizing their outcomes instead optimize the ease of decision making while still getting an acceptable course of action. Most of our biases serve to either allow us to make decisions quicker or minimize the odds of catastrophically bad outcomes for our decisions, which fit nicely with this model. The fact is that indecision is often worse than a bad decision, and the evolutionary forces that shaped our brains are stochastic in nature and thus don't dock points for missed opportunities.

Re: We don't have a hundred biases, we have the wrong model

#24

Post by phkahler » Thu, Jul 21, 2022, 8:47 PM UTC

toomim wrote: post #13

I wrote a PhD dissertation that made this point in 2013, and proposed a new "helocentric" economic model. The key shift is to move the utility function from evaluating a future state of the world to evaluating the utility of an opportunity for attention in the present moment . All the "cognitive errors" that we humans make are with respect to predicting the future. But we all know what we find appealing in the presen…

>> All the "cognitive errors" that we humans make are with respect to predicting the future. But we all know what we find appealing in the present moment.

I like to say that most human problems are a result of the conflict between short and long term goals. This is true at all levels from individuals to small groups, companies, and states. Many, many "failures" can be framed this way. I would say it's not even a problem of predicting the future (thought that is an issue) but of failure to prioritize the future over the present.

Re: We don't have a hundred biases, we have the wrong model

#25

Post by beefman » Thu, Jul 21, 2022, 8:48 PM UTC

No mention of ergodicity economics, which resolves a lot of this. Reinforcement learning was mentioned, which resolves all or nearly all of it.

Re: We don't have a hundred biases, we have the wrong model

#26

Post by twawaaay » Thu, Jul 21, 2022, 8:54 PM UTC

We have some wrong models but this does not mean we don't have biases.

The simplest way to disprove it:

If we did not have biases but wrong models, fixing models would make us unbiased.

But that works very rarely in real life.

Re: We don't have a hundred biases, we have the wrong model

#27

Post by bobthechef » Thu, Jul 21, 2022, 9:11 PM UTC

JoshCole wrote: post #16

If you apply the cognitive biases model to algorithms which have superhuman performance in various games - like AlphaZero, DeepBlue, Pluribus, and so on - the natural result is to conclude that these models are predictably irrational. The reason you get this conclusion is because it turns out to be necessary to trade off theoretical optimal answers for the sake of speed. The behavioral economic view of human irration…

It sound like you're talking about, or at least brushing up against, prudential judgement[0]. Sometimes, the optimal move is not to seek the optimum.

An obvious class of problems is where determining the optimum takes more time than the lifetime of the problem. Say you need to write an algorithm at work that does X, and you need X by tomorrow. If it would take you a week to find the theoretical optimum, then the optimum in a "global" sense is to deliver the best you can within the constraints, not the abstract theoretical optimum. The time to produce the solution is part of the total cost. An imprudent person would either say it's not possible, or never deliver the solution in time.

[0] https://www.newadvent.org/cathen/12517b.htm

Re: We don't have a hundred biases, we have the wrong model

#28

Post by snovv_crash » Thu, Jul 21, 2022, 9:17 PM UTC

JoshCole wrote: post #16

If you apply the cognitive biases model to algorithms which have superhuman performance in various games - like AlphaZero, DeepBlue, Pluribus, and so on - the natural result is to conclude that these models are predictably irrational. The reason you get this conclusion is because it turns out to be necessary to trade off theoretical optimal answers for the sake of speed. The behavioral economic view of human irration…

The thing about continuous space solutions is that they are typically differentiable, which means you can use a gradient descent or LM optimization rather than needing to fully explore the solution space. Typically there are large regions which are heuristically excludable, which is what you are getting at I think, but even an unbiased sampling plus gradient descent often makes problems much more tractable than discrete problems.

Re: We don't have a hundred biases, we have the wrong model

#29

Post by spacebanana7 » Thu, Jul 21, 2022, 9:43 PM UTC

jjk166 wrote: post #23

The rational actor model assumes that a person will behave optimally - using all information available to make and carry out the best decision possible for their goals. I strongly suspect that a better model is that people instead of optimizing their outcomes instead optimize the ease of decision making while still getting an acceptable course of action. Most of our biases serve to either allow us to make decisions q…

The idea you’re describing sounds similar to Satisficing Theory [1]. I agree this approach does a much better job of describing real life decision making than the traditional rational actor model. Unfortunately, Satisficing rarely gets discussed (at least in my experience) in mainstream economics/psychology, despite having been around since the 1950s.

[1] https://en.wikipedia.org/wiki/Satisficing

Re: We don't have a hundred biases, we have the wrong model

#30

Post by Barrin92 » Thu, Jul 21, 2022, 9:54 PM UTC

JoshCole wrote: post #16

If you apply the cognitive biases model to algorithms which have superhuman performance in various games - like AlphaZero, DeepBlue, Pluribus, and so on - the natural result is to conclude that these models are predictably irrational. The reason you get this conclusion is because it turns out to be necessary to trade off theoretical optimal answers for the sake of speed. The behavioral economic view of human irration…

>why this isn't more commonly known among people who call themselves rationalists

because most of these people do nothing else but writing blogs about rationalism. Same reason university tests are sometimes so removed from practicality compared to evaluation criteria in business, the people who make them do nothing else but write these tests.

I suspect if you put some rationalists into the trenches in the Donbass for a week they'd quickly have a more balanced view of what's needed to solve a problem besides rational contemplation.