James L. Carroll: September 2007

I have made some changes and added some stuff...

Web Page Navigation

The menu was giving me fits, the first one wouldn’t go away when printing, so I got a pure CSS menu, but it was popping up a dialog box telling me that I had to purchase it in IE, but not in Firefox, so I didn’t even know about it for a while. Finally I have one that isn’t as pretty, but that I built. It isn’t working perfectly yet, so let me know if you know how to get it to double nest. . . anyway. . .

CS Research

Two new research papers added to the publications page http://james.jlcarroll.net/publications/:

James L. Carroll and Kevin D. Seppi, "No-Free-Lunch and Bayesian Optimality," in Meta-Learning IJCNN Workshop 2007.
We take a Bayesian approach to the issues of bias, meta bias, transfer, overfit, and No-Free-Lunch in the context of supervised learning. If we accept certain relationships between the function class, on training set data, and off training set data, then a graphical model can be created that represents the supervised learning problem. This graphical model dictates a specific algorithm which will be the “optimal” approach to learning the parameters of any given function representation given the variable relationships. Thus, there is an optimal technique for supervised learning. We reconcile this idea of an optimal technique with the ideas of No-Free-Lunch and show how these ideas relate to the concepts of meta and transfer learning through hierarchical versions of the graphical model.
PDF

James L. Carroll, Christopher K. Monson, and Kevin D. Seppi, "A Bayesian CMAC for High Assurance Supervised Learning, Accepted for Publication in Applications of Neural Networks in High-Assurance Systems, IJCNN Workshop 2007.
We analyze the drawbacks to using ANNs in high assurance systems and propose a solution based upon a Bayesian approach with a specific network topology that can be solved in closed form. The Bayesian approach leads to better answers in the traditional sense, while also allowing us to quantify risk and deal with it in a reasonable manner. We demonstrate this approach on several synthetic functions and the Abalone data set.
PDF

Hobbies

I updated the hobbies section and added the following pages:

Aquarium: http://james.jlcarroll.net/hobbies/aquarium.html

Falconry: http://james.jlcarroll.net/hobbies/falconry.html

Literature (link to my goodreads user for now… http://www.goodreads.com/user/show/162048)

JRR Tolkien: http://james.jlcarroll.net/hobbies/tolkien.html

Star Trek (links to STFHE page)

Other hobbies are still links to “under construction” at this point.

Public Pictures Added:

Aquarium: http://james.jlcarroll.net/pictures/Aquarium/index.html

Private Pictures Added:

Picnic with The Fishers: http://james.jlcarroll.net/picturesPrivate/2007/23%20Picnic/index.html