## Kasen Williams, The Man, The Myth, The Legend.

Before I talk about Kasen Williams, I have to clarify that this comes through the lens of someone who played for Issaquah High School, the crosstown rival of Skyline High School, Mr. Williams Alma Mater. We hated Skyline with a burning passion, for very some very petty reasons, mostly that they kept beating us, with Kasen being the prime inflictor of scholastic athletic pain.

I like movies.

## My Five Year Plan

As a 23 year old college graduate I am often asked where I see myself in 5-10 years. I hate this question.

I’d say two thirds of my frustration when asked this type of question comes from the question being pointless, while the other third comes from some fear of not having a satisfactory answer.

## Organizing Collaboration Data as Streams

I had an idea for something I’d like to make/ see made.

Problem it tries to solve: Too many communication / data-provision platforms.

How it solves it: Allow easy curation and inspection of the communication that drives a project forward.

Here’s some random nonsense about my thinking that I want to share/refine while I’m still learning about Observables.

## My Summer of Code Experience

The summer is over and I am very proud to report that I have accomplished most of my goals I set before starting work with MDAnalysis.

Soldiering on!

## Visualizing Data with t-SNE, part 1

Watching Katie Ledecky beat the field by 11 seconds in the 800M swim at the Olympics has led me to the conclusion that I am out of shape. I will never physically be where Katie Ledecky is, but I can use her work ethic as an inspiration to try a little harder academically. Using the Morning Paper as an example, I am going to try to Ledeckify my life a little bit and work hard at becoming more (intellectually) fit every day.

Today’s paper is ‘Visualizing Data with t-SNE’, and given my lack of expertise, I have decided to split this paper into posts. This post will cover:

## SciPy 2016!

Last week I went to Austin, TX to Scipy2016. I wasn’t sure what to expect. How would people communicate? Would I fit in, what talks would interest me? Fortunately the conference was a huge success. I have came away a far more confident and motivated programmer than when I went in.

## Principal Component Analysis

My next subject for bloggery is Principal Component Analysis (PCA) (its sibling Multidimensional scaling has been left out for a future post, but it is just as special, don’t worry). If I were to give a talk on PCA, the slides would be roughly ordered as follows:

• A very short recap of dimension reduction
• PCA, what it stands for, rough background, history
• Eigenvectors (what are those?!)
• Covariance (Because variance matrix didn’t sound cool enough)
• The very fancy sounding method of Lagrange Multipliers (why they aren’t that hard)
• Explain the PCA Algorithm
• Random Walks: What are they, how are they taken on a configuration space
• Interpreting the results after applying PCA on MD simulation data

## A Note from the Author

In the last blog post I wrote the most common critique I received was that I alienated myself from most of my potential audience. In an email I expressed to my Summer of Code mentor Max Linke my problem:

## Diffusion Maps in Molecular Dynamics Analysis

It occurs to me in my previous post I didn’t thoroughly explain the motivation for dimension reduction in general. When we have this data matrix $X$ with $n$ samples and each sample having $m$ features, this number m can be very large. This data contains information that we want to extract, in the case of molecular dynamics simulations these are parameters describing how the dynamics are occurring. But this data can be features that distinguish faces from others in the dataset, handwritten letters and numbers from other numbers, etc. As it is so eloquently put by Porte and Herbst at Arizona

## Dimension Reduction, a review of a review

Hello! This is my first post moving over to a new site built by wintersmith. Originally I was going to use jekyll pages, but there was an issue with the latest Ruby version not being available for Linux, (maybe macs are better…). I spent way too much time figuring out how to install a markdown plugin that allowed for the inclusion of LaTex. I did this all without realizing I could simply include: