Thu, February 29, 9:00 AM
90 MINUTES
Challenges of Modelling from Aristotle to Deep Learning
What is modelling? Why do we make models? What is called a good model? what are the different approaches to modelling? How far we can go solely based on input/output observation and measurement? What are the sources of uncertainty in modelling? Is it possible to resolve the uncertainty? How can we handle uncertainty within a model? In this talk, I am going to discuss the above questions, and more, about modelling; from very basic ancient questions to challenges of deep learning era.
Professor @ University of Tehran
Babak N Araabi was born in 1969 in Tehran. He has got BS, MS, and PhD degrees in 1992, 1996, and 2001, from Sharif Univ. of Tech., Univ. of Tehran, and Texas A&M Univ., respectively, all in Electrical Engineering. He joined the School of ECE at Univ. of Tehran in 2002, where he is currently a professor and director of Machine Learning and Computational Modelling Lab. His research interests lies within the area of machine learning. He is particularly interested in, prediction, causal representation learning, cognitive modelling, fault prognosis, computer vision, deep neural networks, and modelling under uncertainty. Obtained results by Prof. Araabi and his research team, has been published in more 200 peer reviewed journal papers, and the same number of referred conference papers.