Matches in SemOpenAlex for { <https://semopenalex.org/work/W2296952208> ?p ?o ?g. }
- W2296952208 abstract "We consider the online version of the isotonic regression problem. Given a set of linearly ordered points (e.g., on the real line), the learner must predict labels sequentially at adversarially chosen positions and is evaluated by her total squared loss compared against the best isotonic (non-decreasing) function in hindsight. We survey several standard online learning algorithms and show that none of them achieve the optimal regret exponent; in fact, most of them (including Online Gradient Descent, Follow the Leader and Exponential Weights) incur linear regret. We then prove that the Exponential Weights algorithm played over a covering net of isotonic functions has a regret bounded by $Obig(T^{1/3} log^{2/3}(T)big)$ and present a matching $Omega(T^{1/3})$ lower bound on regret. We provide a computationally efficient version of this algorithm. We also analyze the noise-free case, in which the revealed labels are isotonic, and show that the bound can be improved to $O(log T)$ or even to $O(1)$ (when the labels are revealed in isotonic order). Finally, we extend the analysis beyond squared loss and give bounds for entropic loss and absolute loss." @default.
- W2296952208 created "2016-06-24" @default.
- W2296952208 creator A5001603308 @default.
- W2296952208 creator A5008130119 @default.
- W2296952208 creator A5013618078 @default.
- W2296952208 date "2016-03-14" @default.
- W2296952208 modified "2023-09-27" @default.
- W2296952208 title "Online Isotonic Regression" @default.
- W2296952208 cites W1506313179 @default.
- W2296952208 cites W1511694993 @default.
- W2296952208 cites W1518095400 @default.
- W2296952208 cites W1535258871 @default.
- W2296952208 cites W1538452572 @default.
- W2296952208 cites W1570963478 @default.
- W2296952208 cites W167834439 @default.
- W2296952208 cites W178986290 @default.
- W2296952208 cites W1949314668 @default.
- W2296952208 cites W1968248886 @default.
- W2296952208 cites W1969705577 @default.
- W2296952208 cites W2002258118 @default.
- W2296952208 cites W2016511206 @default.
- W2296952208 cites W2077723394 @default.
- W2296952208 cites W2096830573 @default.
- W2296952208 cites W2098824882 @default.
- W2296952208 cites W2112599888 @default.
- W2296952208 cites W2127441697 @default.
- W2296952208 cites W2148825261 @default.
- W2296952208 cites W2153863985 @default.
- W2296952208 cites W2156386646 @default.
- W2296952208 cites W2159058260 @default.
- W2296952208 cites W2611627047 @default.
- W2296952208 cites W2795929191 @default.
- W2296952208 cites W2949983735 @default.
- W2296952208 cites W2950780363 @default.
- W2296952208 cites W2963763760 @default.
- W2296952208 cites W2964170508 @default.
- W2296952208 cites W648428804 @default.
- W2296952208 hasPublicationYear "2016" @default.
- W2296952208 type Work @default.
- W2296952208 sameAs 2296952208 @default.
- W2296952208 citedByCount "3" @default.
- W2296952208 countsByYear W22969522082017 @default.
- W2296952208 countsByYear W22969522082020 @default.
- W2296952208 crossrefType "posted-content" @default.
- W2296952208 hasAuthorship W2296952208A5001603308 @default.
- W2296952208 hasAuthorship W2296952208A5008130119 @default.
- W2296952208 hasAuthorship W2296952208A5013618078 @default.
- W2296952208 hasConcept C105795698 @default.
- W2296952208 hasConcept C114614502 @default.
- W2296952208 hasConcept C118615104 @default.
- W2296952208 hasConcept C126322002 @default.
- W2296952208 hasConcept C134306372 @default.
- W2296952208 hasConcept C138885662 @default.
- W2296952208 hasConcept C14036430 @default.
- W2296952208 hasConcept C151376022 @default.
- W2296952208 hasConcept C17418463 @default.
- W2296952208 hasConcept C185429906 @default.
- W2296952208 hasConcept C2780388253 @default.
- W2296952208 hasConcept C33923547 @default.
- W2296952208 hasConcept C34388435 @default.
- W2296952208 hasConcept C41895202 @default.
- W2296952208 hasConcept C48921125 @default.
- W2296952208 hasConcept C50817715 @default.
- W2296952208 hasConcept C71924100 @default.
- W2296952208 hasConcept C77056095 @default.
- W2296952208 hasConcept C77553402 @default.
- W2296952208 hasConcept C78458016 @default.
- W2296952208 hasConcept C86803240 @default.
- W2296952208 hasConceptScore W2296952208C105795698 @default.
- W2296952208 hasConceptScore W2296952208C114614502 @default.
- W2296952208 hasConceptScore W2296952208C118615104 @default.
- W2296952208 hasConceptScore W2296952208C126322002 @default.
- W2296952208 hasConceptScore W2296952208C134306372 @default.
- W2296952208 hasConceptScore W2296952208C138885662 @default.
- W2296952208 hasConceptScore W2296952208C14036430 @default.
- W2296952208 hasConceptScore W2296952208C151376022 @default.
- W2296952208 hasConceptScore W2296952208C17418463 @default.
- W2296952208 hasConceptScore W2296952208C185429906 @default.
- W2296952208 hasConceptScore W2296952208C2780388253 @default.
- W2296952208 hasConceptScore W2296952208C33923547 @default.
- W2296952208 hasConceptScore W2296952208C34388435 @default.
- W2296952208 hasConceptScore W2296952208C41895202 @default.
- W2296952208 hasConceptScore W2296952208C48921125 @default.
- W2296952208 hasConceptScore W2296952208C50817715 @default.
- W2296952208 hasConceptScore W2296952208C71924100 @default.
- W2296952208 hasConceptScore W2296952208C77056095 @default.
- W2296952208 hasConceptScore W2296952208C77553402 @default.
- W2296952208 hasConceptScore W2296952208C78458016 @default.
- W2296952208 hasConceptScore W2296952208C86803240 @default.
- W2296952208 hasLocation W22969522081 @default.
- W2296952208 hasOpenAccess W2296952208 @default.
- W2296952208 hasPrimaryLocation W22969522081 @default.
- W2296952208 hasRelatedWork W143635698 @default.
- W2296952208 hasRelatedWork W1483114294 @default.
- W2296952208 hasRelatedWork W1611398386 @default.
- W2296952208 hasRelatedWork W1871026852 @default.
- W2296952208 hasRelatedWork W1913735094 @default.
- W2296952208 hasRelatedWork W1934532084 @default.
- W2296952208 hasRelatedWork W2162243185 @default.
- W2296952208 hasRelatedWork W2495622490 @default.