FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890870050> ?p ?o ?g. }

• W2890870050 endingPage "10179" @default.
• W2890870050 startingPage "10169" @default.
• W2890870050 abstract "The problem of learning-to-learn (LTL) or meta-learning is gaining increasing attention due to recent empirical evidence of its effectiveness in applications. The goal addressed in LTL is to select an algorithm that works well on tasks sampled from a meta-distribution. In this work, we consider the family of algorithms given by a variant of Ridge Regression, in which the regularizer is the square distance to an unknown mean vector. We show that, in this setting, the LTL problem can be reformulated as a Least Squares (LS) problem and we exploit a novel meta- algorithm to efficiently solve it. At each iteration the meta-algorithm processes only one dataset. Specifically, it firstly estimates the stochastic LS objective function, by splitting this dataset into two subsets used to train and test the inner algorithm, respectively. Secondly, it performs a stochastic gradient step with the estimated value. Under specific assumptions, we present a bound for the generalization error of our meta-algorithm, which suggests the right splitting parameter to choose. When the hyper-parameters of the problem are fixed, this bound is consistent as the number of tasks grows, even if the sample size is kept constant. Preliminary experiments confirm our theoretical findings, highlighting the advantage of our approach, with respect to independent task learning." @default.
• W2890870050 created "2018-09-27" @default.
• W2890870050 creator A5034260726 @default.
• W2890870050 creator A5037096925 @default.
• W2890870050 creator A5048960946 @default.
• W2890870050 creator A5052978360 @default.
• W2890870050 date "2018-01-01" @default.
• W2890870050 modified "2023-09-24" @default.
• W2890870050 title "Learning To Learn Around A Common Mean" @default.
• W2890870050 cites W1587813652 @default.
• W2890870050 cites W1939941161 @default.
• W2890870050 cites W1978164656 @default.
• W2890870050 cites W1994616650 @default.
• W2890870050 cites W2065180801 @default.
• W2890870050 cites W2121358800 @default.
• W2890870050 cites W2124231679 @default.
• W2890870050 cites W2144752499 @default.
• W2890870050 cites W2149563148 @default.
• W2890870050 cites W2160218441 @default.
• W2890870050 cites W2551277100 @default.
• W2890870050 cites W2552956826 @default.
• W2890870050 cites W2604763608 @default.
• W2890870050 cites W2610857016 @default.
• W2890870050 cites W2753160622 @default.
• W2890870050 cites W2790068046 @default.
• W2890870050 cites W2913340405 @default.
• W2890870050 cites W2951749289 @default.
• W2890870050 cites W2963306862 @default.
• W2890870050 cites W2963424430 @default.
• W2890870050 cites W2963647836 @default.
• W2890870050 cites W2963649256 @default.
• W2890870050 cites W2964308881 @default.
• W2890870050 cites W99485931 @default.
• W2890870050 hasPublicationYear "2018" @default.
• W2890870050 type Work @default.
• W2890870050 sameAs 2890870050 @default.
• W2890870050 citedByCount "36" @default.
• W2890870050 countsByYear W28908700502019 @default.
• W2890870050 countsByYear W28908700502020 @default.
• W2890870050 countsByYear W28908700502021 @default.
• W2890870050 countsByYear W28908700502022 @default.
• W2890870050 crossrefType "proceedings-article" @default.
• W2890870050 hasAuthorship W2890870050A5034260726 @default.
• W2890870050 hasAuthorship W2890870050A5037096925 @default.
• W2890870050 hasAuthorship W2890870050A5048960946 @default.
• W2890870050 hasAuthorship W2890870050A5052978360 @default.
• W2890870050 hasConcept C11413529 @default.
• W2890870050 hasConcept C126255220 @default.
• W2890870050 hasConcept C134306372 @default.
• W2890870050 hasConcept C154945302 @default.
• W2890870050 hasConcept C162324750 @default.
• W2890870050 hasConcept C177148314 @default.
• W2890870050 hasConcept C187736073 @default.
• W2890870050 hasConcept C199360897 @default.
• W2890870050 hasConcept C2777027219 @default.
• W2890870050 hasConcept C2780451532 @default.
• W2890870050 hasConcept C2781002164 @default.
• W2890870050 hasConcept C33923547 @default.
• W2890870050 hasConcept C41008148 @default.
• W2890870050 hasConcept C77553402 @default.
• W2890870050 hasConceptScore W2890870050C11413529 @default.
• W2890870050 hasConceptScore W2890870050C126255220 @default.
• W2890870050 hasConceptScore W2890870050C134306372 @default.
• W2890870050 hasConceptScore W2890870050C154945302 @default.
• W2890870050 hasConceptScore W2890870050C162324750 @default.
• W2890870050 hasConceptScore W2890870050C177148314 @default.
• W2890870050 hasConceptScore W2890870050C187736073 @default.
• W2890870050 hasConceptScore W2890870050C199360897 @default.
• W2890870050 hasConceptScore W2890870050C2777027219 @default.
• W2890870050 hasConceptScore W2890870050C2780451532 @default.
• W2890870050 hasConceptScore W2890870050C2781002164 @default.
• W2890870050 hasConceptScore W2890870050C33923547 @default.
• W2890870050 hasConceptScore W2890870050C41008148 @default.
• W2890870050 hasConceptScore W2890870050C77553402 @default.
• W2890870050 hasLocation W28908700501 @default.
• W2890870050 hasOpenAccess W2890870050 @default.
• W2890870050 hasPrimaryLocation W28908700501 @default.
• W2890870050 hasRelatedWork W2117539524 @default.
• W2890870050 hasRelatedWork W2121358800 @default.
• W2890870050 hasRelatedWork W2237537322 @default.
• W2890870050 hasRelatedWork W2601450892 @default.
• W2890870050 hasRelatedWork W2604763608 @default.
• W2890870050 hasRelatedWork W2742093937 @default.
• W2890870050 hasRelatedWork W2753160622 @default.
• W2890870050 hasRelatedWork W2795900505 @default.
• W2890870050 hasRelatedWork W2923966100 @default.
• W2890870050 hasRelatedWork W2925446857 @default.
• W2890870050 hasRelatedWork W2962928691 @default.
• W2890870050 hasRelatedWork W2963341924 @default.
• W2890870050 hasRelatedWork W2963649256 @default.
• W2890870050 hasRelatedWork W2963775850 @default.
• W2890870050 hasRelatedWork W2964078140 @default.
• W2890870050 hasRelatedWork W2970814018 @default.
• W2890870050 hasRelatedWork W3006977545 @default.
• W2890870050 hasRelatedWork W3007684729 @default.
• W2890870050 hasRelatedWork W3104240813 @default.
• W2890870050 hasRelatedWork W99485931 @default.
• W2890870050 hasVolume "31" @default.

• next