Applications of machine learning and artificial intelligence in marine science is the latest themed set in ICES Journal of Marine Science. Machine learning is an application of artificial intelligence that aims to provide system with the ability to automatically learn from experience without the need for human intervention. Machine (or deep) learning approaches have great potential to improve the quality and extent of marine research by identifying latent patterns and hidden trends, particularly in large datasets, that are intractable using other approaches, e.g. manual investigations carried out by marine experts.
New sensor technology makes it possible to collect large amounts of data from the marine environment. Moored buoys now support long-term monitoring and high resolution measurements of physical, chemical, and biological variables, as well as acoustics and their data is transmitted in real-time via satellite uplink or cabled connection to shore. Sophisticated and heavily instrumented towed observation platforms and autonomous drones are collecting large volumes of different types of data. Human capacity to filter, curate, and analyse all of these data is, however, limited.
Rapidly developing machine learning methods are able to solve very complex, nonlinear, high-dimensional problems as well as process noisy and incomplete data. Deep learning has already supported impressive advances by changing the way that experts analyse and interpret data, as well as in the amount of data that can be processed rapidly.
This themed set of articles brings together contributions on the broad theme of the applications of machine learning and advanced data systems to research, monitor, and manage marine organisms and ecosystems. The chosen papers provide as wide a selection as possible of case studies that demonstrate the applications, utility, and promise of machine learning in marine science – the selected papers make clear that there is great potential for machine learning to contribute to rapid advances in the field.
The latest issue of ICES Journal of Marine Science features two themed sets, Science in support of a nonlinear non-equilibrium world and Applications of machine learning and artificial intelligence in marine science. Both themed sets are free to access for all readers.
Cartoon by Bas Köhler. Click to enlarge.