An introduction to distributed training of deep neural networks for segmentation tasks with large seismic datasets.
Claire BirnieHaithem JarrayaFredrik HansteenPublished in: CoRR (2021)
Keyphrases
- neural network
- training process
- segmentation method
- training algorithm
- segmentation algorithm
- image segmentation
- distributed systems
- amazon mechanical turk
- level set
- feedforward neural networks
- medical images
- feed forward neural networks
- deep architectures
- distributed environment
- training dataset
- data mining tasks
- ground truth labels
- pattern recognition
- multi agent
- fully automatic
- backpropagation algorithm
- text classification tasks
- cooperative
- training examples
- benchmark datasets
- self organizing maps
- multi layer perceptron
- segmentation accuracy
- mobile agents
- object segmentation
- recurrent neural networks
- energy function
- neural network training
- training and testing data
- multiscale
- seismic data
- recurrent networks
- computational grids
- object detection
- training phase
- supervised learning