Our work on measures of divergence using representation entropy was accepted to ICASSP 2022. Congrats to Jhoan Keider and Oscar on their first publication in our lab. The pre print is available here: https://arxiv.org/abs/2112.01583
Our lab's PI has been awarded funding by the Department of Defense for a new project through its Defense Established Program to Stimulate Competitive Research (DEPSCoR) awards competition. Nathan Jacobs, associate professor in the Department of Computer Science, will serve as co-investigator and senior mentor.
This project will develop the theoretical foundations of alternative definitions of entropy and mutual information to address a fundamental problem in Machine Learning: building objective functions that can handle and integrate data from multiple sources with minimal supervision.
The CLIP Lab welcomes Santiago Posso Murillo and Jhoan Keider Hoyos Osorio to Lexington.
Originally from Colombia, Santiago and Jhoan Keider will pursue MS and PhD degrees in Electrical Engineering at UK.
We are excited to having you here.
Nick's research will focus on developing a method of feature extraction using a novel definition of
mutual information. This method will allow the construction of models with multiple levels of processing and nonlinearities and the untangling of feature dependencies.
Phillip will work on a mechanincal device to simulate the process of image acquisition performed by the eye, which combines information from the retina captured over time and eye motion processes.
This device will be used to obtain data to test different algorithmic strategies for information extraction that combine visual and motor modalities.
