Jorge F. Silva
Professor of Electrical Engineering
University of Chile
About me
I am faculty working at the Information and Decision System group, Universidad de Chile, Santiago, Chile. Throughout my research career, I have been passionate about the interaction between learning and information theory. Some of my recent works have been on: universal source coding (lossy and lossless) in $\infty$-alphabets, the estimation of information measures (entropy, mutual information and divergence), learning and decision with data-rate constraints, and the characterization of nearly sparse ($\ell_p$-compressible) random sequences when relaxing stationary and ergodic assumptions.
I am currently interested in analyzing some connections between learning and information theory; learning and universal source coding; sparsity, invariance and other statistical structures that can be learned-detected from the data; as well as understanding the generalization capacity of complex learning algorithms.
I love music (the aspects of blues and jazz improvisation) and playing electric guitar.
Interests
- Statistical Learning
- Statistical Signal Processing
- Source Coding
- Sparse Models
- Information Theory
- Probability Theory
Education
-
Ph. D. in Electrical Engineering, 2009
University of Southern California
-
M. Sc. in Electrical Engineering, 2005
University of Southern California
-
B.Sc in Electrical Engineering, 2000
Universidad de Chile
