On strong asymptotic uniform smoothness and convexity
نویسندگان
چکیده
منابع مشابه
Strong Martingale Type and Uniform Smoothness
Abstract. We introduce stronger versions of the usual notions of martingale type p ≤ 2 and cotype q ≥ 2 of a Banach space X and show that these concepts are equivalent to uniform p-smoothness and q-convexity, respectively. All these are metric concepts, so they depend on the particular norm in X. These concepts allow us to get some more insight into the fine line between X being isomorphic to a...
متن کاملOn the duality of strong convexity and strong smoothness: Learning applications and matrix regularization
We show that a function is strongly convex with respect to some norm if and only if its conjugate function is strongly smooth with respect to the dual norm. This result has already been found to be a key component in deriving and analyzing several learning algorithms. Utilizing this duality, we isolate a single inequality which seamlessly implies both generalization bounds and online regret bou...
متن کاملRSG: Beating Subgradient Method without Smoothness and Strong Convexity
In this paper, we study the efficiency of a Restarted SubGradient (RSG) method that periodically restarts the standard subgradient method (SG). We show that, when applied to a broad class of convex optimization problems, RSG method can find an ǫ-optimal solution with a low complexity than SG method. In particular, we first show that RSG can reduce the dependence of SG’s iteration complexity on ...
متن کاملPOWER TYPE UNIFORM CONVEXITY OF X VIA p - ASYMPTOTIC UNIFORM CONVEXITY OF L r ( X
We show that if Lr(X), 1 < r < ∞, has an asymptotically uniformly convex renorming of power type then X admits a uniformly convex norm of power type.
متن کاملApplications of strong convexity--strong smoothness duality to learning with matrices
It is known that a function is strongly convex with respect to some norm if and only if its conjugate function is strongly smooth with respect to the dual norm. This result has already been found to be a key component in deriving and analyzing several learning algorithms. Utilizing this duality, we isolate a single inequality which seamlessly implies both generalization bounds and online regret...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Revista Matemática Complutense
سال: 2017
ISSN: 1139-1138,1988-2807
DOI: 10.1007/s13163-017-0246-1