Machine learning designates the statistical methods that learn to perform a task (classify elements, predict values, etc.) by inferring properties and patterns from a set of known data (training set) and then can perform the same task to provide predictions on a new set of data (test set). These methods have the potential to considerably accelerate the processing of the massive amount of data collected in marine science, particularly for the fisheries science process.
The objectives WGMLEARN are to:
- review the current applications of machine learning in marine science as well as the new developments in machine learning that would be of potential interest
- identify key challenges
- inform data collection and storage to enable the use of machine learning on marine data archives
- identify trends and future needs, to promote the use of relevant machine learning technologies
If you are interested in this working group, contact ICES Secretariat.
