Richard Creamer's Personal Website

Welcome to my personal website summarizing my interests, projects, and prior work examples.

Tagline: I am a 'needle' :-)

Introduction

I have worn software engineer/architect, entrepreneur, software manager, product manager, and principal investigator hats in multiple, diverse fields, including many key startup roles. I enjoy and excel at entering/innovating in new domains. Currently based in Northern California.

Interests
• Machine Learning, C.S., AI, optimization
• Artificial general intelligence
• Stereoscopic imaging/graphics
• Semantics
• Augmented reality (AR, VR)
• Web-scale analytics
• Electronic Health Records
• Knowledge representation and browsing
• Computational argumentation
• Graphs (node-edge)
• Next-generation social networking platforms
Work & Education

Employment
• Innovest Systems, Software Engineer, 2014 - present
• Cool Apps Software, Software Engineer, 2012 - present
Education, University
• University of California, Davis, B.S. Applied Physics
Education, Online (Coursera.org)

Other profiles/pages:

I rarely update this website anymore, but below are mostly older, historical updates...
• Python kNN classification exercise [Oct-2016]
• Currently (Fall-2016) studying Python's Machine Learning libraries:
• Numpy + SciKit-Learn
•  Here is a plot I coded using Matplotlib and a data set from one of the Python ML books I'm studying
• Mean Absolute Deviation in Python/Numpy - my StackOverflow answer
• This looks to be a good introduction to Deep Learning (URL)
• By invitation and for the 2nd time, I will again be a Community TA for the Stanford online Machine Learning course beginning Sept. 22, 2014 (Coursera).
• Kaggle.com Higgs Boson Challenge
• The purpose of this challenge was to identify a particular Higgs Boson decay event from background noise data.
• Below is a histogram of the predicted Real output numbers from an early Random Forest classifier.  The best threshold on these Real outputs was found by applying the quantile() R function on the Train set labels.
• A few more details are here.
• Project Euler Solutions (coding examples):
• I just added a new page w/solutions to a few selected Project Euler problems.  PLEASE do not use this code on ProjectEuler.net !
• Weight Lifting Exercises Dataset (Summary Page)
• I recently built a Random Forest classifier in R which classifies a human's exercise activity from 52 accelerometer and tilt/gryo sensor measurements.  The training set had about 19,000 labeled samples.
• CodeEval Challenge: A quick C hack got me a 99.841% ranking on a CodeEval.com programming challenge:-)  PLEASE do not use my code on CodeEval.com !
• Recent online ML/Data Science courses:
• Completed 3 online courses from the Johns Hopkins Data Science specialization series on Data Science:
• Certificates of completion and course topic summaries can be viewed here
• I just learned about and joined CodeEval.com, looking forward to finding time to solve some of the interesting challenges on this site!
• I just progressed to Level 2 on ProjectEuler.net (top 7.26%) by completing 50+ problems.  Update: I added a new page w/ selected solutions/code.  Screenshot showing advancement to Level 2:
• Concluded role as online Stanford Machine Learning course Community TA! (part-time) - last day of course: May 26, 2014
Slide decks (Click left link to go to Slide Decks page)

Disclaimer:
Any comments, ideas, IP, or opinions I make on the Web are entirely my own, and are completely distinct from, do not reflect upon, or represent in any way, my employers, past, present, and future.

Selected Site Highlights...

Autogeny Message-Based RPC (early thinking

Autogeny's 2005 approach to modeling three types of temporal properties in node-edge graphs (similar to RDF triples)

Thumbnail screen from a C# image processing app
I wrote a couple of years ago (my photos)

Matplotlib plot I coded from Python ML book data set

RGB Histogram - C# 14-bits per channel, alpha-blending

4-Tuple Graph Data, Metadata, Vocabulary Model

C++ Qt OpenGL Rendering of Mt. Shasta