FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208162305> ?p ?o ?g. }

• W3208162305 abstract "Most research on novel techniques for 3D Medical Image Segmentation (MIS) is currently done using Deep Learning with GPU accelerators. The principal challenge of such technique is that a single input can easily cope computing resources, and require prohibitive amounts of time to be processed. Distribution of deep learning and scalability over computing devices is an actual need for progressing on such research field. Conventional distribution of neural networks consist in “data parallelism”, where data is scattered over resources (e.g., GPUs) to parallelize the training of the model. However, “experiment parallelism” is also an option, where different training processes (i.e., on a hyper-parameter search) are parallelized across resources. While the first option is much more common on 3D image segmentation, the second provides a pipeline design with less dependence among parallelized processes, allowing overhead reduction and more potential scalability. In this work we present a design for distributed deep learning training pipelines, focusing on multi-node and multi-GPU environments, where the two different distribution approaches are deployed and benchmarked. We take as proof of concept the 3D U-Net architecture, using the MSD Brain Tumor Segmentation dataset, a state-of-art problem in medical image segmentation with high computing and space requirements. Using the BSC MareNostrum supercomputer as benchmarking environment, we use TensorFlow and Ray as neural network training and experiment distribution platforms. We evaluate the experiment speed-up when parallelizing, showing the potential for scaling out on GPUs and nodes. Also comparing the different parallelism techniques, showing how experiment distribution leverages better such resources through scaling, e.g. by a speed-up factor from x12 to x14 using 32 GPUs. Finally, we provide the implementation of the design open to the community, and the non-trivial steps and methodology for adapting and deploying a MIS case as the here presented." @default.
• W3208162305 created "2021-11-08" @default.
• W3208162305 creator A5008356835 @default.
• W3208162305 creator A5013177206 @default.
• W3208162305 creator A5063484640 @default.
• W3208162305 creator A5083087205 @default.
• W3208162305 date "2022-05-01" @default.
• W3208162305 modified "2023-09-26" @default.
• W3208162305 title "Distributing Deep Learning Hyperparameter Tuning for 3D Medical Image Segmentation" @default.
• W3208162305 cites W1901129140 @default.
• W3208162305 cites W1984020445 @default.
• W3208162305 cites W2007146250 @default.
• W3208162305 cites W2057947771 @default.
• W3208162305 cites W2069533927 @default.
• W3208162305 cites W2106411961 @default.
• W3208162305 cites W2432481613 @default.
• W3208162305 cites W2464708700 @default.
• W3208162305 cites W2621776666 @default.
• W3208162305 cites W2799277203 @default.
• W3208162305 cites W2883265831 @default.
• W3208162305 cites W2884700152 @default.
• W3208162305 cites W2899792733 @default.
• W3208162305 cites W2915126261 @default.
• W3208162305 cites W2947510367 @default.
• W3208162305 cites W2953410600 @default.
• W3208162305 cites W2962881747 @default.
• W3208162305 cites W2963276418 @default.
• W3208162305 cites W2964054038 @default.
• W3208162305 cites W2964121744 @default.
• W3208162305 cites W2970009439 @default.
• W3208162305 cites W2972291421 @default.
• W3208162305 cites W3017243633 @default.
• W3208162305 cites W3036218681 @default.
• W3208162305 cites W3037640857 @default.
• W3208162305 cites W3043787446 @default.
• W3208162305 cites W3048581364 @default.
• W3208162305 cites W3113588636 @default.
• W3208162305 cites W3128822558 @default.
• W3208162305 cites W3154814778 @default.
• W3208162305 doi "https://doi.org/10.1109/ipdpsw55747.2022.00172" @default.
• W3208162305 hasPublicationYear "2022" @default.
• W3208162305 type Work @default.
• W3208162305 sameAs 3208162305 @default.
• W3208162305 citedByCount "2" @default.
• W3208162305 countsByYear W32081623052022 @default.
• W3208162305 countsByYear W32081623052023 @default.
• W3208162305 crossrefType "proceedings-article" @default.
• W3208162305 hasAuthorship W3208162305A5008356835 @default.
• W3208162305 hasAuthorship W3208162305A5013177206 @default.
• W3208162305 hasAuthorship W3208162305A5063484640 @default.
• W3208162305 hasAuthorship W3208162305A5083087205 @default.
• W3208162305 hasBestOaLocation W32081623052 @default.
• W3208162305 hasConcept C108583219 @default.
• W3208162305 hasConcept C111919701 @default.
• W3208162305 hasConcept C113775141 @default.
• W3208162305 hasConcept C119857082 @default.
• W3208162305 hasConcept C124504099 @default.
• W3208162305 hasConcept C154945302 @default.
• W3208162305 hasConcept C173608175 @default.
• W3208162305 hasConcept C199360897 @default.
• W3208162305 hasConcept C2779960059 @default.
• W3208162305 hasConcept C2781172179 @default.
• W3208162305 hasConcept C41008148 @default.
• W3208162305 hasConcept C43521106 @default.
• W3208162305 hasConcept C48044578 @default.
• W3208162305 hasConcept C50644808 @default.
• W3208162305 hasConcept C61483411 @default.
• W3208162305 hasConcept C77088390 @default.
• W3208162305 hasConcept C83283714 @default.
• W3208162305 hasConcept C89600930 @default.
• W3208162305 hasConceptScore W3208162305C108583219 @default.
• W3208162305 hasConceptScore W3208162305C111919701 @default.
• W3208162305 hasConceptScore W3208162305C113775141 @default.
• W3208162305 hasConceptScore W3208162305C119857082 @default.
• W3208162305 hasConceptScore W3208162305C124504099 @default.
• W3208162305 hasConceptScore W3208162305C154945302 @default.
• W3208162305 hasConceptScore W3208162305C173608175 @default.
• W3208162305 hasConceptScore W3208162305C199360897 @default.
• W3208162305 hasConceptScore W3208162305C2779960059 @default.
• W3208162305 hasConceptScore W3208162305C2781172179 @default.
• W3208162305 hasConceptScore W3208162305C41008148 @default.
• W3208162305 hasConceptScore W3208162305C43521106 @default.
• W3208162305 hasConceptScore W3208162305C48044578 @default.
• W3208162305 hasConceptScore W3208162305C50644808 @default.
• W3208162305 hasConceptScore W3208162305C61483411 @default.
• W3208162305 hasConceptScore W3208162305C77088390 @default.
• W3208162305 hasConceptScore W3208162305C83283714 @default.
• W3208162305 hasConceptScore W3208162305C89600930 @default.
• W3208162305 hasFunder F4320320300 @default.
• W3208162305 hasLocation W32081623051 @default.
• W3208162305 hasLocation W32081623052 @default.
• W3208162305 hasLocation W32081623053 @default.
• W3208162305 hasOpenAccess W3208162305 @default.
• W3208162305 hasPrimaryLocation W32081623051 @default.
• W3208162305 hasRelatedWork W1547595128 @default.
• W3208162305 hasRelatedWork W1595151633 @default.
• W3208162305 hasRelatedWork W2948658236 @default.
• W3208162305 hasRelatedWork W2960184797 @default.
• W3208162305 hasRelatedWork W2972093541 @default.
• W3208162305 hasRelatedWork W3115553566 @default.

• next