Recent work has shown that AdaBoost can be viewed as an algorithm that maximizes the margin on the training data via functional gradient descent. Under this interpretation, the weight computed by AdaBoost, for each hypothesis generated, can be viewed as a step size parameter in a gradient descent search. Friedman...
A comparison and analysis of jitter for five different architectures of ring oscillators using a novel simulation technique developed by Professor Forbes' group is presented. Ring oscillators have become an essential building block in many digital and synchronous communications system due to their integrated nature and are widely used in...
This thesis work evaluates the need for a re-configurable cross compiler for the X32V processor architecture and discusses the process of developing a cross compiler for X32V. X32V is a new processor intended at the embedded applications domain whose instruction set is designed based on the widely used MIPS processor....