Documentation

Books.ConvexAnalysis_Rockafellar_1970.Chapters.Chap02.section09_part12

theorem closure_posHomogeneousGenerated_infimum_rightScalarMultiple {n : } {f f0_plus : (Fin n)EReal} (hfclosed : ClosedConvexFunction f) (hfproper : ProperConvexFunctionOn Set.univ f) (hf0 : f 0 > 0) (hrec : (epigraph Set.univ f).recessionCone = epigraph Set.univ f0_plus) :
have k := positivelyHomogeneousConvexFunctionGenerated f; ProperConvexFunctionOn Set.univ k (∀ (x : Fin n), convexFunctionClosure k x = sInf (Set.insert (f0_plus x) {z : EReal | ∃ (lam : ), 0 < lam z = rightScalarMultiple f lam x})) (∀ (x : Fin n), zSet.insert (f0_plus x) {z : EReal | ∃ (lam : ), 0 < lam z = rightScalarMultiple f lam x}, convexFunctionClosure k x = z) (0 effectiveDomain Set.univ fClosedConvexFunction k ∀ (x : Fin n), k x = sInf {z : EReal | ∃ (lam : ), 0 < lam z = rightScalarMultiple f lam x})

Theorem 9.7. Let f be a closed proper convex function on ℝ^n with f(0) > 0, and let k be the positively homogeneous convex function generated by f. Then k is proper and (convexFunctionClosure k)(x) is the infimum of (f λ)(x) over λ > 0 together with λ = 0^+ (encoded here as f0_plus), the infimum being attained for each x. If 0 ∈ dom f, then k is closed and the λ = 0^+ term can be omitted (but the infimum might not be attained).