Abstract. Since the onset of Reservoir Computing (RC) in the early 2000s, it has proven itself to be a productive paradigm for training and understanding Recurrent Neural Networks and beyond. As someone fortunate enough to be in the field from quite early on, I would like to start with a short introduction to RC for newcomers, and share some of my perspectives on RC in a broader and changing context of (Deep) Machine Learning, as well as some perspectives that RC sheds on the broader related fields, in this overview talk.
Dr. Lukoševičius is an associate professor at the Kaunas University of Technology, Faculty of Informatics. He has been one of the pioneers of Reservoir Computing, having done his Master’s and PhD degrees with Prof. Herbert Jaeger in the MINDS research group at Jacobs University Bremen (now Constructor University). Besides Reservoir Computing, his current research is more broadly related to machine learning, biomedicine, image, natural language, and signal processing and analysis. Read more on his personal website.