Felipe Tobar is a CONICYeeT Research Fellow in Machine Learning and Signal Processing working in the Center for Mathematical Modeling at 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.