Parametric Learning in Collaborative Signal Processing
Mar 06, 2012
from 10:00 AM to 12:00 PM
|Where||ENGR. IV Bldg. Faraday Room 67-124|
|Add event to calendar||
Advisor: Kung Yao
The advances in embedded computing and communication technology have emerged to foster the genesis of a new digital era that embraces a plethora of innovative applications. Central to the realization of these immersive innovations is to exploit the functions that bridge the cyber space and the physical world. The topics on parametric learning examine these instrumental functions and require novel perspectives on the corresponding efficacy. We address the problems of learning the fundamental parameters in the likelihood functions, detectors, probability distributions and classifiers related to many applications. The objective of our study is to unveil the underlying parametric functions by harnessing collaborative processing.
Our first topic entails a unified framework of learning the parameters in the affine functions that appear in directional operating and sensing modalities. We pursue a robust solution set to recover the affine functions by taking the advantage of structured collaboration. In the second topic, we address the detection performance of cognitive computing in the presence of interference. Given the constraints on the interference power, we quantify the tight upper and lower bounds on the probabilistic detection metrics and characterize the interference wall. The third topic pertains to the initialization of random-finite-set posterior density functions. A spatially distributed particle filter is derived to update and aggregate the weights and particles in a multi-Bernoulli set. The fourth topic involves distributed hierarchy formation and support vector machines. We offer the solutions to clustering and classification that quantitatively achieve high clustering balance and classification fairness.
Biography:Juo-Yu Lee is currently a PhD candidate in the Electrical Engineering Department. His recent research focuses on collaborative learning and array signal processing. He has been the recipient of Garmin Scholarship and Rundel Fellowship.