Felipe Tobar is a Research Fellow in Machine Learning in the Center for Mathematical Modeling and Adjunct Lecturer at the Department of Mathematical Engineering, Universidad de Chile. Between Sept. 2014 and July 2015, he was a Research Associate at the Machine Learning Group, University of Cambridge. Felipe holds a PhD in Adaptive Signal Processing from Imperial College London (2014), and BSc (2006) and MSc (2010) degrees in Electrical Engineering from the Universidad de Chile.
His current research interests include:
- Desing and identification of covariance functions and spectrum of time series using Bayesian nonparametrics.
- Multidimensional extensions of reproducing kernel Hilbert spaces (complex, quaternion, vector and multi-kernel) as a basis for multivariate adaptive signal estimation and prediction.
- Bayesian model identification and design using kernel learning
- Applications on: wind prediction, body sensor trajectory tracking, chaotic signal prediction, financial signals and point-of-gaze estimation.