Matches in SemOpenAlex for { <https://semopenalex.org/work/W1947123202> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W1947123202 abstract "This thesis considers two online matrix learning problems: the problem of online Principle Component Analysis (PCA) and the problem of learning rotations online. Previous papers proposed two online algorithms for these problems, which are the Matrix Exponentiated Gradient (MEG) algorithm and the Gradient Descent (GD) algorithm, respectively. This thesis evaluates these two algorithms by their regret, which is the additional total loss of the online algorithm over the total loss of the best offline comparator (chosen in hindsight).In Chapter 2, we show that for online PCA, both algorithms achieve essentially the same and optimal (within a constant factor) regret bound when the bound is expressed as a function of the number of trials. However, when considering regret bounds as functions of the loss of the best comparator, MEG remains optimal and strictly outperforms GD. Chapter 3 considers a generalization of online PCA in which we use the combination of the compression gain of PCA and the cosine similarity to measure the closeness between two directions (i.e. unit length vectors). Such a combined measurement involves both the first and second moments of the learner's randomized prediction, and therefore, we propose an online algorithm that maintains both a vector and a matrix as its parameter simultaneously. In particular, in each trial of the game, this algorithm uses a novel iterative procedure to decompose its parameter pair into a mixture of at most $n$ pure directions. Chapter 4 considers the problem of learning rotations online. We show that for this problem, the GD algorithm is optimal (up to a constant) for both the regret bounds that are functions of the number of trials and the regret bounds that are functions of the loss of best comparator." @default.
- W1947123202 created "2016-06-24" @default.
- W1947123202 creator A5077127148 @default.
- W1947123202 date "2015-01-01" @default.
- W1947123202 modified "2023-09-27" @default.
- W1947123202 title "Optimal Online Learning with Matrix Parameters" @default.
- W1947123202 hasPublicationYear "2015" @default.
- W1947123202 type Work @default.
- W1947123202 sameAs 1947123202 @default.
- W1947123202 citedByCount "0" @default.
- W1947123202 crossrefType "journal-article" @default.
- W1947123202 hasAuthorship W1947123202A5077127148 @default.
- W1947123202 hasConcept C102408133 @default.
- W1947123202 hasConcept C105795698 @default.
- W1947123202 hasConcept C106487976 @default.
- W1947123202 hasConcept C11413529 @default.
- W1947123202 hasConcept C126255220 @default.
- W1947123202 hasConcept C134306372 @default.
- W1947123202 hasConcept C153258448 @default.
- W1947123202 hasConcept C154945302 @default.
- W1947123202 hasConcept C159985019 @default.
- W1947123202 hasConcept C192562407 @default.
- W1947123202 hasConcept C196921405 @default.
- W1947123202 hasConcept C33923547 @default.
- W1947123202 hasConcept C41008148 @default.
- W1947123202 hasConcept C50644808 @default.
- W1947123202 hasConcept C50817715 @default.
- W1947123202 hasConcept C77553402 @default.
- W1947123202 hasConceptScore W1947123202C102408133 @default.
- W1947123202 hasConceptScore W1947123202C105795698 @default.
- W1947123202 hasConceptScore W1947123202C106487976 @default.
- W1947123202 hasConceptScore W1947123202C11413529 @default.
- W1947123202 hasConceptScore W1947123202C126255220 @default.
- W1947123202 hasConceptScore W1947123202C134306372 @default.
- W1947123202 hasConceptScore W1947123202C153258448 @default.
- W1947123202 hasConceptScore W1947123202C154945302 @default.
- W1947123202 hasConceptScore W1947123202C159985019 @default.
- W1947123202 hasConceptScore W1947123202C192562407 @default.
- W1947123202 hasConceptScore W1947123202C196921405 @default.
- W1947123202 hasConceptScore W1947123202C33923547 @default.
- W1947123202 hasConceptScore W1947123202C41008148 @default.
- W1947123202 hasConceptScore W1947123202C50644808 @default.
- W1947123202 hasConceptScore W1947123202C50817715 @default.
- W1947123202 hasConceptScore W1947123202C77553402 @default.
- W1947123202 hasLocation W19471232021 @default.
- W1947123202 hasOpenAccess W1947123202 @default.
- W1947123202 hasPrimaryLocation W19471232021 @default.
- W1947123202 hasRelatedWork W1483952878 @default.
- W1947123202 hasRelatedWork W1771507533 @default.
- W1947123202 hasRelatedWork W1920773645 @default.
- W1947123202 hasRelatedWork W2068373302 @default.
- W1947123202 hasRelatedWork W2149433471 @default.
- W1947123202 hasRelatedWork W2186453173 @default.
- W1947123202 hasRelatedWork W2491144192 @default.
- W1947123202 hasRelatedWork W2546057486 @default.
- W1947123202 hasRelatedWork W28319768 @default.
- W1947123202 hasRelatedWork W2945414017 @default.
- W1947123202 hasRelatedWork W2963252079 @default.
- W1947123202 hasRelatedWork W2963812988 @default.
- W1947123202 hasRelatedWork W2964055417 @default.
- W1947123202 hasRelatedWork W2964069971 @default.
- W1947123202 hasRelatedWork W2988115836 @default.
- W1947123202 hasRelatedWork W3000941013 @default.
- W1947123202 hasRelatedWork W3046712426 @default.
- W1947123202 hasRelatedWork W3156692118 @default.
- W1947123202 hasRelatedWork W3175012888 @default.
- W1947123202 hasRelatedWork W3180361179 @default.
- W1947123202 isParatext "false" @default.
- W1947123202 isRetracted "false" @default.
- W1947123202 magId "1947123202" @default.
- W1947123202 workType "article" @default.