I mean, Bayes and Popper, they’re not like night and day, right?
There are some stark differences.
#Popperian claim that positive justification is impossible.
#Induction doesn’t exist (or at least , matter in science)
#Popper was prepared to consider the existence of Propensities objective.probabilities, whereas Bayesians, particularly those who follow Jaynes believe in determinism and subjective probability.
#Popperian refutation is all or nothing, whereas Bayesian negative information is gradual.
#In Popperism, there can be more than one front running or most favoured theory, even after the falsifiable ones have been falsified, since there aren’t quantifiable degrees of confirmation.
*Explanation
For Popper and Deutsch, theories need to be explanatory, not just predictive. Bayesian confirmation and disconfirmation only target prediction directly—if they are achieving explanation or ontological correspondence , that would be the result of a convenient coincidence.
#Conjectures.
For Popperians, the construction of good theoretical conjectures is as important as testing them. Bayesian seem quite uninterested in where hypotheses come from.
#Simplicity
For Deutschians, being hard-to-vary is the preferred principle of parsimony. For Yudkowskians, it’s computation complexity.
#Error correction
For Popperians us something you actually do.
Popperians like to put forward hypotheses that are easy to refute. Bayesians approve theoretically of “updating”, but dislike objections and criticisms in practice.
#(Long term) prediction is basically impossible .
More Deutsch than Popper—DD believed that the growth and unpredictability of knowledge . The creation of knowledge is so unpredictable and radical that long term predictions cannot be made. Often summarised to “prediction is impossible”. Of course , Bayesians are all about prediction—but the predictive power of Ates tends only to be demonstrated in you models, where the ontology isn’t changing under your feet. Their AI I predictions are explicitly intuition based.
#Optimism versus Doom.
Deutsch is highly optimistic that continuing knowledge creation will change the world for the better (a kind of moral realism is a component of this). Yudkowsky thinks advanced AI is our last invention and will kill us all.
Which is that Popperianism bottoms out onto common sense.
Falsification and fallbiilism are quite intuitive to scientists … on the other hand, both ideas took some time to arrive...they weren’t obvious to Aristotle or Bacon.
The non existence of induction is not common sense.
So I, I don’t really have such a thing as changing my mind because the state of my mind is always, it’s a playing field of different hypotheses, right? I always have a group of hypotheses and there’s never one that it’s like, oh this is my mind on this one. Every time I make a prediction, I actually have all the different hypotheses weigh in, weighted by their probability, and thy all make the prediction together.”
What’s the difference? Is updating is spectral, changing your mind is binary?
I mean, Solomonov induction does grow its knowledge and grow its predictive confidence, right
It starts off with omniscience, in the sense of possible hypotheses and then gets whittled down.
one of the good Bayesian critiques about frequentism that I like. So I, we, I totally agree with you. That, that, that the world is deterministic, non stochastic, and randomness doesn’t actually occur in nature. I, I agree.
Determinism is not a fact.
Liron Shapira: we, or we might we, we, there’s, there’s just no epistemic value to treating the universe as ontologically fundamentally, non deterministic, and the strongest example I’ve seen of that is in quantum theory, like the idea that a quantum collapses. ontologically fundamental to
There’s always epistemological value in believing the truth. If the universe is not deterministic,a rationalist should want to believe so.
What I’m saying is probability is not the best tool to reason about the future precisely because the future is chaotic and unpredictable, right?
Liron:
There are some stark differences.
#Popperian claim that positive justification is impossible.
#Induction doesn’t exist (or at least , matter in science)
#Popper was prepared to consider the existence of Propensities objective.probabilities, whereas Bayesians, particularly those who follow Jaynes believe in determinism and subjective probability.
#Popperian refutation is all or nothing, whereas Bayesian negative information is gradual.
#In Popperism, there can be more than one front running or most favoured theory, even after the falsifiable ones have been falsified, since there aren’t quantifiable degrees of confirmation.
*Explanation
For Popper and Deutsch, theories need to be explanatory, not just predictive. Bayesian confirmation and disconfirmation only target prediction directly—if they are achieving explanation or ontological correspondence , that would be the result of a convenient coincidence.
#Conjectures.
For Popperians, the construction of good theoretical conjectures is as important as testing them. Bayesian seem quite uninterested in where hypotheses come from.
#Simplicity
For Deutschians, being hard-to-vary is the preferred principle of parsimony. For Yudkowskians, it’s computation complexity.
#Error correction
For Popperians us something you actually do.
Popperians like to put forward hypotheses that are easy to refute. Bayesians approve theoretically of “updating”, but dislike objections and criticisms in practice.
#(Long term) prediction is basically impossible .
More Deutsch than Popper—DD believed that the growth and unpredictability of knowledge . The creation of knowledge is so unpredictable and radical that long term predictions cannot be made. Often summarised to “prediction is impossible”. Of course , Bayesians are all about prediction—but the predictive power of Ates tends only to be demonstrated in you models, where the ontology isn’t changing under your feet. Their AI I predictions are explicitly intuition based.
#Optimism versus Doom.
Deutsch is highly optimistic that continuing knowledge creation will change the world for the better (a kind of moral realism is a component of this). Yudkowsky thinks advanced AI is our last invention and will kill us all.
Falsification and fallbiilism are quite intuitive to scientists … on the other hand, both ideas took some time to arrive...they weren’t obvious to Aristotle or Bacon.
The non existence of induction is not common sense.
What’s the difference? Is updating is spectral, changing your mind is binary?
It starts off with omniscience, in the sense of possible hypotheses and then gets whittled down.
Determinism is not a fact.
There’s always epistemological value in believing the truth. If the universe is not deterministic,a rationalist should want to believe so.
Depends on whether it’s near or far.