I am sedated

I Am A Home Page

Student+Misc Info

4th year PhD student @ CMU-SCS

Enrolled with the CNBC training program in the joint Pitt/CMU Center for the Neural Basis of Cognition

For some Cortex code and Matlab routines to deal with Cortex data files, click here.

For code that computes expectations of maximum entropy for spiking neurons and finite data length, check out this code . The paper reference will be given upon journal acceptance.

Check out this animation of the transformation matrix learned using unsupervised info-max on artificial image data.

For details on the above, check out the paper we did for class. Working on additions and extensions.

Treasurer of the GSA.

The proposed budget for Fiscal 1999-2000.

The Constitution of Senate/GSA, which describes the processes for joint Senate/GSA funding meetings

Family photos

The archetypal antagonist

My bookmarks

Ex-Job Info

Grad student @ CMU, but here's what I used to do before that...

Programming and Research work for the Imaging Systems Lab, Robotics Department, Carnegie Mellon University.

Past projects: Optical Chinese and Hangul (Korean) Character recognition, Graphical Image Analysis, plus other small things, (including getting a real home page together)

Random info

Name: Rick Romero
Phone: 412-268-2580 (CMU-CLU0)
Phone: 412-268-6414
Email: rickr+@cmu.edu
Favorite Drugs: nicotine and caffeine
Goal in Life: To one day have no goals

Some Source code

Splay tree code: This is an implementation of Tarjan and Sleator's splay tree algorithm, a self-adjusting binary search tree. I'm working on docs but the code is pretty easy to use. It should compile with just about any ANSI C compiler. Amortized log(n) performance.
Rectangle Access Trees: This is an implementation of a suite of routines for dealing with rectangles in a 2D plane. Why rectangles? Well, it's nice if you are doing image manipulation, it is a natural way of representing image blobs or regions, and that's the one I managed to find a nice algorithm for. The problem was that I needed dynamic insertion and deletion in O(log(n)) time, plus I wanted to do range queries in sub-linear time. (Well, in the worst case it's obviously not possible, since a range query can return everything in your tree, but the time spent limiting the search is sub-linear.) For a description of the algorithm, check out the paper. This one actually has some man pages and other misc. documentation. This set of software includes the Splay tree code above, and makes heavy use of it.
Combinatorial graphs code: Some code that embeds graphs into the combinatorial form. I did this for an algorithms class, so it may or may not be useful to you. Allows you to get the genus of a graph, and compute the connected components.

If it is illegal to operate a Ponzi scheme, can we force the government via lawsuit to operate Social Security as a true trust fund, and not just another pyramid scheme? Better yet, can we force Congress through the courts to operate on a perpetual balanced budget? Worth looking into, seeing as how the balanced budget amendment looks like it will never pass. Isn't it about time this country elected a president who was a leader?

Random pointers

Imaging Systems Lab home page (where I worked)
CMU's Contributed home page
CS Tech Report Dbase

Back door to CMU

CIA Server Cause you never know when you might need a spy...