Interested in AI alignment and safety, abstractions, generalisation, and multi-agent systems.
https://profiles.sussex.ac.uk/p555273-fernando-rosas, https://profiles.imperial.ac.uk/f.rosas/about
Fernando Rosas
Karma: 34
The point is that a coarse-graining can turn a quasi-probability into a probability. This is significant because quasi-probabilities are limited in that they cannot be sampled etc.
Does that make sense?
Great post! I have a small question about the sophisticated chooser. I see that it was clarified that it can potentially be money-pumped. Does that mean that it doesn’t satisfy dynamical consistency?
What about consequentialism? Is that depending on if the local preferences depend or not on closed branches? (is that the definition of “locality”?)