Regarding curriculum learning: I think its very neglected, and seems likely to be a core component of prosaic alignment approaches. The idea of a “basin of attraction for corrigibility (or other desirable properties)” seems likely to rely on appropriate choice of curriculum.
Regarding curriculum learning: I think its very neglected, and seems likely to be a core component of prosaic alignment approaches. The idea of a “basin of attraction for corrigibility (or other desirable properties)” seems likely to rely on appropriate choice of curriculum.