I follow most of the post, but got confused by the non-cyclic graph & halting problem/computational intractability part.
Is the non-cyclic graph a way of modelling causality as state transitions?
Re: computational intractability; I understand your argument as saying:
Dualism is used as a fundamental interpretative tool, since it makes it tractable to analyze feedback loops between brain and environment.
When meditating, dualism gets broken down due to lack of feedback loops and an increase in neutral annealing, leading to nondual world models
Questions:
A. Why is the awakened brain capable of performing “computationally intractable” computation (non-dual & feedback loopy)? Seems more like an aptness thing than a fundamental impossibility thing (a la halting problem)
B. Does all awakening-inducing meditative practices involve cutting feedback loops? Is awakening a theravada/Samantha thing?
“Dualism is used as a fundamental interpretative tool” is a good way of putting it.
“When meditating, dualism gets broken down due to lack of feedback loops and an increase in neutral annealing, leading to nondual world models” ← Also yes.
What I was trying to get at with the non-cyclic graph stuff is that if all the brain did was non-causally model an external world, then it is always possible in principle to create a simulation where sensory inputs affect the simulation unidirectionally, and information flows through the simulation itself with no cycles. This is like how our weather prediction computers do not predictably affect the weather itself, or how planetariums do not affect the movement of the stars.
Embedded world optimizers are different. Your world model affect your motor outputs which affect the physical world which affects the world model. This is a cycle. In this way, non-embedded world optimizers such as chess engines (which are effectively stateless due to Minimax) differ from embedded world optimizers such as our brains, because embedded world optimizers cause cyclic causal loops when interacting with their environment. It’s basically the Time Travel Paradox problem, except without any need for time travel. Your brain makes sense of this by scribbling “free will” over a link in the chain it doesn’t want to look too closely at. By pretending this link is effectively random, it simplifies what appears to be an intractable self-referential problem into what appears to be (but isn’t) a tractable non-self-referential one. Meditation temporarily breaks a link in the cycle by stopping the motor outputs where you make changes to the physical external world. Without that link in the cycle, the loop is cut and the causality becomes non-cyclic [while you’re meditating].
Answer A: The computational intractability is a side effect of being an embedded world optimizer 𝓪 𝓫𝓪𝓵𝓵 𝓸𝓯 𝓷𝓮𝓾𝓻𝓸𝓷𝓼 that must satisfy the conflicting optimization targets of “create a simulated world that accurately models external physical reality” and “optimize the simulated world into a desired state”. The Gordian knot of computational intractability is transcended by giving up trying to be an interpretible world optimizer as defined by, say, decision theory. All those abstractions that insulate your value system from your world model? You just throw them out. The computationally intractable problem is still computationally intractable. It just stops being a problem for you, because you’re not trying to solve that problem anymore.
Answer B: Awakening is not limited to therevada/Samatha. I got there via Zen, for example, which is non-Therevadan. You can even detect Awakening in non-Buddhists when you know what to look for. For example, I believe Mary Baker Eddy ended up at stream entry via Christian prayer.
Is the non-cyclic graph a way of modelling causality as state transitions?
I hope I answered your question. I’m uncertain what you’re trying to ask here due to an ambiguity in the word “state”. The word “state”, can refer to many different things in this context, including altered states of consciousness (as distinct from altered traits), the computational concept of state (vs statelessness), and also attractors (which are similar to altered traits).
I follow most of the post, but got confused by the non-cyclic graph & halting problem/computational intractability part.
Is the non-cyclic graph a way of modelling causality as state transitions?
Re: computational intractability; I understand your argument as saying:
Dualism is used as a fundamental interpretative tool, since it makes it tractable to analyze feedback loops between brain and environment.
When meditating, dualism gets broken down due to lack of feedback loops and an increase in neutral annealing, leading to nondual world models
Questions: A. Why is the awakened brain capable of performing “computationally intractable” computation (non-dual & feedback loopy)? Seems more like an aptness thing than a fundamental impossibility thing (a la halting problem) B. Does all awakening-inducing meditative practices involve cutting feedback loops? Is awakening a theravada/Samantha thing?
Thank you for the excellent comment.
“Dualism is used as a fundamental interpretative tool” is a good way of putting it.
“When meditating, dualism gets broken down due to lack of feedback loops and an increase in neutral annealing, leading to nondual world models” ← Also yes.
What I was trying to get at with the non-cyclic graph stuff is that if all the brain did was non-causally model an external world, then it is always possible in principle to create a simulation where sensory inputs affect the simulation unidirectionally, and information flows through the simulation itself with no cycles. This is like how our weather prediction computers do not predictably affect the weather itself, or how planetariums do not affect the movement of the stars.
Embedded world optimizers are different. Your world model affect your motor outputs which affect the physical world which affects the world model. This is a cycle. In this way, non-embedded world optimizers such as chess engines (which are effectively stateless due to Minimax) differ from embedded world optimizers such as our brains, because embedded world optimizers cause cyclic causal loops when interacting with their environment. It’s basically the Time Travel Paradox problem, except without any need for time travel. Your brain makes sense of this by scribbling “free will” over a link in the chain it doesn’t want to look too closely at. By pretending this link is effectively random, it simplifies what appears to be an intractable self-referential problem into what appears to be (but isn’t) a tractable non-self-referential one. Meditation temporarily breaks a link in the cycle by stopping the motor outputs where you make changes to the physical external world. Without that link in the cycle, the loop is cut and the causality becomes non-cyclic [while you’re meditating].
Answer A: The computational intractability is a side effect of being
an embedded world optimizer𝓪 𝓫𝓪𝓵𝓵 𝓸𝓯 𝓷𝓮𝓾𝓻𝓸𝓷𝓼 that must satisfy the conflicting optimization targets of “create a simulated world that accurately models external physical reality” and “optimize the simulated world into a desired state”. The Gordian knot of computational intractability is transcended by giving up trying to be an interpretible world optimizer as defined by, say, decision theory. All those abstractions that insulate your value system from your world model? You just throw them out. The computationally intractable problem is still computationally intractable. It just stops being a problem for you, because you’re not trying to solve that problem anymore.Answer B: Awakening is not limited to therevada/Samatha. I got there via Zen, for example, which is non-Therevadan. You can even detect Awakening in non-Buddhists when you know what to look for. For example, I believe Mary Baker Eddy ended up at stream entry via Christian prayer.
I hope I answered your question. I’m uncertain what you’re trying to ask here due to an ambiguity in the word “state”. The word “state”, can refer to many different things in this context, including altered states of consciousness (as distinct from altered traits), the computational concept of state (vs statelessness), and also attractors (which are similar to altered traits).