Convex function Optimization Foundations
noun phrase
Definition: A function whose value at any point on the line segment between two points in its domain is no greater than the weighted average of the function’s values at those two points (Jensen-type convexity condition) [Boyd, Vandenberghe 2004].
Example in context: “The sum of KL divergences is a strictly convex function, which attains its global minimum at z(cid:63), so will strictly increase as we advance along the ray.” [Piliouras et al. 2022]
Related terms: strictly convex function, concave function (contrast), convex objective