Deep learning drives the so-called third wave of artificial intelligence (AI) and has dominating many automate tasks. However, what is deep learning? Why is it called “deep” learning? This webinar will present a general introduction to deep learning models, i.e., from artificial neural networks (ANN) to deep neural networks (DNN). Three major types of deep learning models will be covered, including convolution neural networks (CNN) mainly used in computer vision, recurrent neural networks (RNN) mainly targeting at time-series problems, and the latest Transformer that can be used in both computer vision and time-series analysis. Their extensions, e.g., different CNN architectures, RNN extensions (GRU, LSTM), Transformer based (BERT for language modelling and DETR, ViT for computer vision) and their typical applications will also be introduced. Finally, a general pipeline of applying deep learning in practical problems will be presented.
This course is appropriate for general audience, both developers and data scientists with basic knowledge about deep learning.
Dr. Junyong You
Senior Researcher at Norwegian Research Centre AS. He has a wide range of R&D experience with data analysis, artificial intelligence, computer vision and applications in multimedia, energy, aquaculture, energy, etc. His research interests include machine learning, semantic multimedia content analysis, visual attention mechanism and modelling, and multimedia quality of experience management. He is the author or co-author of more than 70 scientific articles published in books, premier conferences and journals.