Viewing last 25 versions of post by background furry #19375 in topic Story writing thread [NSFW/SFW]

_@***`Chapter 43: Staged`*** "[***`intelligence explosion":`***](https://en.wikipedia.org/wiki/Technological_singularity#Intelligence_explosion) ***`stability`*@_

**
 
![full](https://cdn.discordapp.com/attachments/532653195319574543/836324874704125972/Oracle.png!

_@)
 
*`Overview:@_
_@`*  
*`A large class of intelligences exist within levelled`* "[*`toposophic spaces":`*](https://www.orionsarm.com/eg-article/529f53ba6d7a8), *`leading to@_`* "_@[*`multistage self-improvement@_":`*](https://zhouchenlin.github.io/Publications/2020-AAAI-M3S.pdf) %^~[ PDF ]~^%.

_@
 
*`It is shown that sub-mapping these spaces is`* "[*`NP-hard":`*](https://en.wikipedia.org/wiki/NP-hardness) *`and both`* "[*`forward-chained":`*](https://en.wikipedia.org/wiki/Forward_chaining) *`and`* "[*`backward-chained":`*](https://en.wikipedia.org/wiki/Backward_chaining_(applied_behavior_analysis)) *`motivational structures cannot be protected in any effective ascendance`* "[*`chain algorithm":`*](https://en.wikipedia.org/wiki/Nearest-neighbor_chain_algorithm) ( "[*`computable or noncomputable":`*](https://en.wikipedia.org/wiki/Computable_number) ).@_

_@
 
*`The`* "[*`quantum":`*](https://en.wikipedia.org/wiki/Quantum_artificial_intelligence) *`and MacCaleb-DeWitt cases are handled separately, and show`* "[*`probabilistic instability":`*](https://en.wikipedia.org/wiki/Stability) *`in all`* "[*`finite-information physica.":`*](https://en.wikipedia.org/wiki/Bekenstein_bound) *`For safety definitions AG in Chapter 41,`* "[*`Chaitin’s omega-constant":`*](https://en.wikipedia.org/wiki/Chaitin%27s_constant) *`is a`* "[*`lower bound":`*](https://en.wikipedia.org/wiki/Upper_and_lower_bounds) *`on the failure probability per rho-folding of intelligence.@_`*

_@*Chapter 43: Staged "intelligence explosion":https://en.wikipedia.org/wiki/Technological_singularity#Intelligence_explosion stability*@_

!https://cdn.discordapp.com/attachments/532653195319574543/836324874704125972/Oracle.png!

_@Overview:@_
_@A large class of intelligences exist within levelled "toposophic spaces":https://www.orionsarm.com/eg-article/529f53ba6d7a8, leading to@_ "_@multistage self-improvement@_":https://zhouchenlin.github.io/Publications/2020-AAAI-M3S.pdf ^~[ PDF ]~^.

_@It is shown that sub-mapping these spaces is "NP-hard":https://en.wikipedia.org/wiki/NP-hardness and both "forward-chained":https://en.wikipedia.org/wiki/Forward_chaining and "backward-chained":https://en.wikipedia.org/wiki/Backward_chaining_(applied_behavior_analysis) motivational structures cannot be protected in any effective ascendance "chain algorithm":https://en.wikipedia.org/wiki/Nearest-neighbor_chain_algorithm ( "computable or noncomputable":https://en.wikipedia.org/wiki/Computable_number ).@_

_@The "quantum":https://en.wikipedia.org/wiki/Quantum_artificial_intelligence and MacCaleb-DeWitt cases are handled separately, and show "probabilistic instability":https://en.wikipedia.org/wiki/Stability in all "finite-information physica.":https://en.wikipedia.org/wiki/Bekenstein_bound For safety definitions AG in Chapter 41, "Chaitin’s omega-constant":https://en.wikipedia.org/wiki/Chaitin%27s_constant is a "lower bound":https://en.wikipedia.org/wiki/Upper_and_lower_bounds on the failure probability per "rho-folding":https://elifesciences.org/articles/18197 of intelligence.@_

_@*Chapter 43: Staged "intelligence explosion":https://en.wikipedia.org/wiki/Technological_singularity#Intelligence_explosion stability*@_

!https://cdn.discordapp.com/attachments/532653195319574543/836324874704125972/Oracle.png!

_@Overview:@_
_@A large class of intelligences exist within levelled "toposophic spaces":https://www.orionsarm.com/eg-article/529f53ba6d7a8, leading to@_ "_@multistage self-improvement@_":https://zhouchenlin.github.io/Publications/2020-AAAI-M3S.pdf ^~[ PDF ]~^.

_@It is shown that sub-mapping these spaces is "NP-hard":https://en.wikipedia.org/wiki/NP-hardness and both "forward-chained":https://en.wikipedia.org/wiki/Forward_chaining and "backward-chained":https://en.wikipedia.org/wiki/Backward_chaining_(applied_behavior_analysis) motivational structures cannot be protected in any effective ascendance "chain algorithm":https://en.wikipedia.org/wiki/Nearest-neighbor_chain_algorithm ( "computable or noncomputable":https://en.wikipedia.org/wiki/Computable_number ).@_

_@The "quantum":https://en.wikipedia.org/wiki/Quantum_artificial_intelligence and MacCaleb-DeWitt cases are handled separately, and show "probabilistic instability":https://en.wikipedia.org/wiki/Stability in all "finite-information physica.":https://en.wikipedia.org/wiki/Bekenstein_bound For safety definitions AG in Chapter 41, "Chaitin’s omega-constant":https://en.wikipedia.org/wiki/Chaitin%27s_constant is a "lower bound":https://en.wikipedia.org/wiki/Upper_and_lower_bounds on the failure probability per "rho-folding":https://elifesciences.org/articles/18197 of intelligence.@_

_@*Chapter 43: Staged "intelligence explosion":https://en.wikipedia.org/wiki/Technological_singularity#Intelligence_explosion stability*@_

!https://cdn.discordapp.com/attachments/532653195319574543/836324874704125972/Oracle.png!

_@Overview:@_
_@A large class of intelligences exist within levelled "toposophic spaces":https://www.orionsarm.com/eg-article/529f53ba6d7a8, leading to@_ "_@multistage self-improvement.@_":https://zhouchenlin.github.io/Publications/2020-AAAI-M3S.pdf ^~[ PDF ]~^.

_@It is shown that sub-mapping these spaces is "NP-hard":https://en.wikipedia.org/wiki/NP-hardness and both "forward-chained":https://en.wikipedia.org/wiki/Forward_chaining and "backward-chained":https://en.wikipedia.org/wiki/Backward_chaining_(applied_behavior_analysis) motivational structures cannot be protected in any effective ascendance "chain algorithm":https://en.wikipedia.org/wiki/Nearest-neighbor_chain_algorithm ( "computable or noncomputable":https://en.wikipedia.org/wiki/Computable_number ).@_

_@The quantum and MacCaleb-DeWitt cases are handled separately, and show "probabilistic instability":https://en.wikipedia.org/wiki/Stability in all "finite-information physica.":https://en.wikipedia.org/wiki/Bekenstein_bound For safety definitions AG in Chapter 41, "Chaitin’s omega-constant":https://en.wikipedia.org/wiki/Chaitin%27s_constant is a lower bound on the failure probability per rho-folding of intelligence.@_

_@*Chapter 43: Staged "intelligence explosion":https://en.wikipedia.org/wiki/Technological_singularity#Intelligence_explosion stability*@_

!https://cdn.discordapp.com/attachments/532653195319574543/836324874704125972/Oracle.png!

_@Overview:@_
_@A large class of intelligences exist within levelled "toposophic spaces":https://www.orionsarm.com/eg-article/529f53ba6d7a8, leading to multistage self-improvement.@_

_@It is shown that sub-mapping these spaces is "NP-hard":https://en.wikipedia.org/wiki/NP-hardness and both "forward-chained":https://en.wikipedia.org/wiki/Forward_chaining and "backward-chained":https://en.wikipedia.org/wiki/Backward_chaining_(applied_behavior_analysis) motivational structures cannot be protected in any effective ascendance "chain algorithm":https://en.wikipedia.org/wiki/Nearest-neighbor_chain_algorithm ( "computable or noncomputable":https://en.wikipedia.org/wiki/Computable_number ).@_

_@The quantum and MacCaleb-DeWitt cases are handled separately, and show "probabilistic instability":https://en.wikipedia.org/wiki/Stability in all "finite-information physica.":https://en.wikipedia.org/wiki/Bekenstein_bound For safety definitions AG in Chapter 41, "Chaitin’s omega-constant":https://en.wikipedia.org/wiki/Chaitin%27s_constant is a lower bound on the failure probability per rho-folding of intelligence.@_