I’m unclear on how the probability distribution over utility functions would be implemented. A complete specification of how to evaluate evidence seems hard to do right. Also, why should we expect we can produce a pool of utility functions that includes an adequate one?
I’m unclear on how the probability distribution over utility functions would be implemented. A complete specification of how to evaluate evidence seems hard to do right. Also, why should we expect we can produce a pool of utility functions that includes an adequate one?