P
Initializing...
Lemma 9, p. 15 — the elevated objective of the relaxation optimum q̃ bounds the objective of every state-dependent policy: Ôbj(q̃) ≥ Opt_m · Prove2Me