The Joint Statistical Seminars are aimed at researchers, employees, and students.
This week's seminar is given by Saikat Chatterjee, School of Electrical Engineering and Computer Science, KTH-Royal Institute of Technology, Sweden.
Title: DANSE - Data-Driven Non-linear State Estimation of Model-free Process in Unsupervised Learning Setup
Abstract: This seminar will address a standard Bayesian state estimation problem, like Kalman Filter. The major new thing is that Kalman Filter to Particle Filter, almost all the methods are assumed to know the underline state space model or process model (also known as process dynamics), but our new method DANSE does not. DANSE learns from noisy measurements without access to clean data and/or state space models. That means DANSE learns in an unsupervised manner from noisy measurements, and fully model-free. It is an interesting combination of deep learning and Bayesian learning, and then perform Bayesian estimation.
Short bio of the speaker: Saikat Chatterjee is an associate professor at the School of Electrical Engineering and Computer Science, KTH-Royal Institute of Technology, Sweden. He has published more than 100 papers in international journals and conferences. He was a co-author of the paper that won the best student paper award at ICASSP 2010. He is currently on the editorial board of the Digital Signal Processing Journal, Elsevier, and EURASIP Journal on Advances in Signal Processing. He was the chair of the EURASIP Special Area Team on Signal and Data Analytics for Machine Learning. He offered a Deep Neural Networks course to Ericsson and other Swedish industries. His current research interests are statistical signal processing, machine learning, medical data analysis, data analytics, and speech, audio and image processing.
