We will go through the basics of transfer learning and self-supervised learning as well as some examples from the AI research at RISE, Research institutes of Sweden.
Deep learning has become the workhorse of modern AI. Deep neural networks power systems that excel in image processing, playing ancient board-games, and interpreting natural language. This has led to a number of break-throughs for application areas such as medical image processing, advanced data analytics, machine translation, and autonomous vehicles.
It can be difficult and expensive to obtain the data needed to train the models. Transfer learning, the process of training a model on a dataset that is different from the target domain where the model will be used and evaluated, is one strategy that can help to overcome that problem.
Olof Mogren is heading the deep learning research group in Gothenburg at RISE. He received his PhD in computer science from Chalmers university of technology in 2018, with a thesis on deep representation learning for natural language processing.
Vi hjälper dig att utvecklas. Träffa andra ingenjörer genom att delta i våra event.