FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048688049> ?p ?o ?g. }

• W3048688049 abstract "Incremental gradient (IG) methods, such as stochastic gradient descent and its variants are commonly used for large scale optimization in machine learning. Despite the sustained effort to make IG methods more data-efficient, it remains an open question how to select a training data subset that can theoretically and practically perform on par with the full dataset. Here we develop CRAIG, a method to select a weighted subset (or coreset) of training data that closely estimates the full gradient by maximizing a submodular function. We prove that applying IG to this subset is guaranteed to converge to the (near)optimal solution with the same convergence rate as that of IG for convex optimization. As a result, CRAIG achieves a speedup that is inversely proportional to the size of the subset. To our knowledge, this is the first rigorous method for data-efficient training of general machine learning models. Our extensive set of experiments show that CRAIG, while achieving practically the same solution, speeds up various IG methods by up to 6x for logistic regression and 3x for training deep neural networks." @default.
• W3048688049 created "2020-08-18" @default.
• W3048688049 creator A5009166220 @default.
• W3048688049 creator A5091272738 @default.
• W3048688049 creator A5091674429 @default.
• W3048688049 date "2019-06-05" @default.
• W3048688049 modified "2023-10-01" @default.
• W3048688049 title "Coresets for Data-efficient Training of Machine Learning Models" @default.
• W3048688049 cites W1409984952 @default.
• W3048688049 cites W1492009297 @default.
• W3048688049 cites W1522301498 @default.
• W3048688049 cites W1533861849 @default.
• W3048688049 cites W1680189815 @default.
• W3048688049 cites W1757790940 @default.
• W3048688049 cites W1836465849 @default.
• W3048688049 cites W1866529865 @default.
• W3048688049 cites W1912128066 @default.
• W3048688049 cites W1939652453 @default.
• W3048688049 cites W1959112989 @default.
• W3048688049 cites W1972711404 @default.
• W3048688049 cites W1980287119 @default.
• W3048688049 cites W1981313592 @default.
• W3048688049 cites W1992208280 @default.
• W3048688049 cites W2029828838 @default.
• W3048688049 cites W2045964207 @default.
• W3048688049 cites W2047152541 @default.
• W3048688049 cites W2061863621 @default.
• W3048688049 cites W2076503082 @default.
• W3048688049 cites W2095531095 @default.
• W3048688049 cites W2096908304 @default.
• W3048688049 cites W2105875671 @default.
• W3048688049 cites W2107438106 @default.
• W3048688049 cites W2113372738 @default.
• W3048688049 cites W2134130436 @default.
• W3048688049 cites W2146502635 @default.
• W3048688049 cites W2162704429 @default.
• W3048688049 cites W2188731843 @default.
• W3048688049 cites W2221583060 @default.
• W3048688049 cites W2549654239 @default.
• W3048688049 cites W2594845788 @default.
• W3048688049 cites W2601251344 @default.
• W3048688049 cites W2884265051 @default.
• W3048688049 cites W2948223045 @default.
• W3048688049 cites W2952215077 @default.
• W3048688049 cites W2962844620 @default.
• W3048688049 cites W2963923734 @default.
• W3048688049 cites W2963933682 @default.
• W3048688049 cites W2964016923 @default.
• W3048688049 cites W2964112534 @default.
• W3048688049 cites W2964332587 @default.
• W3048688049 cites W2970224333 @default.
• W3048688049 cites W2982573148 @default.
• W3048688049 cites W584312462 @default.
• W3048688049 cites W6908809 @default.
• W3048688049 hasPublicationYear "2019" @default.
• W3048688049 type Work @default.
• W3048688049 sameAs 3048688049 @default.
• W3048688049 citedByCount "2" @default.
• W3048688049 countsByYear W30486880492020 @default.
• W3048688049 crossrefType "posted-content" @default.
• W3048688049 hasAuthorship W3048688049A5009166220 @default.
• W3048688049 hasAuthorship W3048688049A5091272738 @default.
• W3048688049 hasAuthorship W3048688049A5091674429 @default.
• W3048688049 hasConcept C111919701 @default.
• W3048688049 hasConcept C112680207 @default.
• W3048688049 hasConcept C119857082 @default.
• W3048688049 hasConcept C121332964 @default.
• W3048688049 hasConcept C126255220 @default.
• W3048688049 hasConcept C14036430 @default.
• W3048688049 hasConcept C145446738 @default.
• W3048688049 hasConcept C153258448 @default.
• W3048688049 hasConcept C153294291 @default.
• W3048688049 hasConcept C154945302 @default.
• W3048688049 hasConcept C162324750 @default.
• W3048688049 hasConcept C177264268 @default.
• W3048688049 hasConcept C178621042 @default.
• W3048688049 hasConcept C199360897 @default.
• W3048688049 hasConcept C206688291 @default.
• W3048688049 hasConcept C2524010 @default.
• W3048688049 hasConcept C2777211547 @default.
• W3048688049 hasConcept C2777303404 @default.
• W3048688049 hasConcept C2984842247 @default.
• W3048688049 hasConcept C33923547 @default.
• W3048688049 hasConcept C41008148 @default.
• W3048688049 hasConcept C50522688 @default.
• W3048688049 hasConcept C50644808 @default.
• W3048688049 hasConcept C51632099 @default.
• W3048688049 hasConcept C68339613 @default.
• W3048688049 hasConcept C78458016 @default.
• W3048688049 hasConcept C86803240 @default.
• W3048688049 hasConceptScore W3048688049C111919701 @default.
• W3048688049 hasConceptScore W3048688049C112680207 @default.
• W3048688049 hasConceptScore W3048688049C119857082 @default.
• W3048688049 hasConceptScore W3048688049C121332964 @default.
• W3048688049 hasConceptScore W3048688049C126255220 @default.
• W3048688049 hasConceptScore W3048688049C14036430 @default.
• W3048688049 hasConceptScore W3048688049C145446738 @default.
• W3048688049 hasConceptScore W3048688049C153258448 @default.
• W3048688049 hasConceptScore W3048688049C153294291 @default.
• W3048688049 hasConceptScore W3048688049C154945302 @default.

• next