|
|
|
|
|
|
by AIorNot
340 days ago
|
|
|
I also wonder deeply the philosophy behind imagination of neural models What are the implications of relying on deep networks for instantiating and running the abstractions we usually hand upon physics and transistors Is this a type of VM Is an imagined VM Turing complete? |
|
|
Fascinating question. My “vibe” opinion is that it is, but there are limits on the meaning of Turing completeness that do not apply within traditional computing paradigms, vis a vis scaling costs. My intuition is that scaling costs in imaginary VMs would be quadratic rather that linear, e.g. a task that takes twice the memory takes ^2 compute instead.