FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3036743188> ?p ?o ?g. }

• W3036743188 abstract "Hyper-parameter optimization is crucial for pushing the accuracy of a deep learning model to its limits. A hyper-parameter optimization job, referred to as a study, involves numerous trials of training a model using different training knobs, and therefore is very computation-heavy, typically taking hours and days to finish. We observe that trials issued from hyper-parameter optimization algorithms often share common hyper-parameter sequence prefixes. Based on this observation, we propose Hippo, a hyper-parameter optimization system that removes redundancy in the training process to reduce the overall amount of computation significantly. Instead of executing each trial independently as in existing hyper-parameter optimization systems, Hippo breaks down the hyper-parameter sequences into stages and merges common stages to form a tree of stages (called a stage-tree), then executes a stage once per tree on a distributed GPU server environment. Hippo is applicable to not only single studies, but multi-study scenarios as well, where multiple studies of the same model and search space can be formulated as trees of stages. Evaluations show that Hippo's stage-based execution strategy outperforms trial-based methods such as Ray Tune for several models and hyper-parameter optimization algorithms, reducing GPU-hours and end-to-end training time significantly." @default.
• W3036743188 created "2020-06-25" @default.
• W3036743188 creator A5013247222 @default.
• W3036743188 creator A5041084728 @default.
• W3036743188 creator A5081591481 @default.
• W3036743188 creator A5083084972 @default.
• W3036743188 date "2020-06-21" @default.
• W3036743188 modified "2023-09-23" @default.
• W3036743188 title "Hippo: Taming Hyper-parameter Optimization of Deep Learning with Stage Trees" @default.
• W3036743188 cites W1480850417 @default.
• W3036743188 cites W1922655562 @default.
• W3036743188 cites W2108598243 @default.
• W3036743188 cites W2131241448 @default.
• W3036743188 cites W2176412452 @default.
• W3036743188 cites W2193413348 @default.
• W3036743188 cites W2194775991 @default.
• W3036743188 cites W2525778437 @default.
• W3036743188 cites W2622263826 @default.
• W3036743188 cites W2625885504 @default.
• W3036743188 cites W2732547613 @default.
• W3036743188 cites W2746314669 @default.
• W3036743188 cites W2749581528 @default.
• W3036743188 cites W2772678505 @default.
• W3036743188 cites W2785873347 @default.
• W3036743188 cites W2790089582 @default.
• W3036743188 cites W2883265831 @default.
• W3036743188 cites W2895709428 @default.
• W3036743188 cites W2899071864 @default.
• W3036743188 cites W2919594608 @default.
• W3036743188 cites W2921935051 @default.
• W3036743188 cites W2946822591 @default.
• W3036743188 cites W2949117887 @default.
• W3036743188 cites W2949676527 @default.
• W3036743188 cites W2962760235 @default.
• W3036743188 cites W2963037989 @default.
• W3036743188 cites W2963065629 @default.
• W3036743188 cites W2963334428 @default.
• W3036743188 cites W2963341956 @default.
• W3036743188 cites W2963390429 @default.
• W3036743188 cites W2963577254 @default.
• W3036743188 cites W2963587345 @default.
• W3036743188 cites W2963702144 @default.
• W3036743188 cites W2963815651 @default.
• W3036743188 cites W2964054038 @default.
• W3036743188 cites W2964108773 @default.
• W3036743188 cites W2964121744 @default.
• W3036743188 cites W2970971581 @default.
• W3036743188 cites W2987954639 @default.
• W3036743188 cites W3010637564 @default.
• W3036743188 cites W3022298203 @default.
• W3036743188 cites W3037833767 @default.
• W3036743188 cites W3118608800 @default.
• W3036743188 cites W6908809 @default.
• W3036743188 doi "https://doi.org/10.48550/arxiv.2006.11972" @default.
• W3036743188 hasPublicationYear "2020" @default.
• W3036743188 type Work @default.
• W3036743188 sameAs 3036743188 @default.
• W3036743188 citedByCount "0" @default.
• W3036743188 crossrefType "posted-content" @default.
• W3036743188 hasAuthorship W3036743188A5013247222 @default.
• W3036743188 hasAuthorship W3036743188A5041084728 @default.
• W3036743188 hasAuthorship W3036743188A5081591481 @default.
• W3036743188 hasAuthorship W3036743188A5083084972 @default.
• W3036743188 hasBestOaLocation W30367431881 @default.
• W3036743188 hasConcept C105795698 @default.
• W3036743188 hasConcept C111919701 @default.
• W3036743188 hasConcept C113174947 @default.
• W3036743188 hasConcept C11413529 @default.
• W3036743188 hasConcept C119857082 @default.
• W3036743188 hasConcept C134306372 @default.
• W3036743188 hasConcept C154945302 @default.
• W3036743188 hasConcept C33923547 @default.
• W3036743188 hasConcept C41008148 @default.
• W3036743188 hasConcept C45374587 @default.
• W3036743188 hasConcept C73586568 @default.
• W3036743188 hasConcept C98045186 @default.
• W3036743188 hasConceptScore W3036743188C105795698 @default.
• W3036743188 hasConceptScore W3036743188C111919701 @default.
• W3036743188 hasConceptScore W3036743188C113174947 @default.
• W3036743188 hasConceptScore W3036743188C11413529 @default.
• W3036743188 hasConceptScore W3036743188C119857082 @default.
• W3036743188 hasConceptScore W3036743188C134306372 @default.
• W3036743188 hasConceptScore W3036743188C154945302 @default.
• W3036743188 hasConceptScore W3036743188C33923547 @default.
• W3036743188 hasConceptScore W3036743188C41008148 @default.
• W3036743188 hasConceptScore W3036743188C45374587 @default.
• W3036743188 hasConceptScore W3036743188C73586568 @default.
• W3036743188 hasConceptScore W3036743188C98045186 @default.
• W3036743188 hasLocation W30367431881 @default.
• W3036743188 hasOpenAccess W3036743188 @default.
• W3036743188 hasPrimaryLocation W30367431881 @default.
• W3036743188 hasRelatedWork W2354062721 @default.
• W3036743188 hasRelatedWork W2766697009 @default.
• W3036743188 hasRelatedWork W2961085424 @default.
• W3036743188 hasRelatedWork W3046775127 @default.
• W3036743188 hasRelatedWork W4225307033 @default.
• W3036743188 hasRelatedWork W4285260836 @default.
• W3036743188 hasRelatedWork W4286629047 @default.
• W3036743188 hasRelatedWork W4306321456 @default.
• W3036743188 hasRelatedWork W4306674287 @default.

• next