October 31, 2017
It’s been quite some time since my last post. My silence hasn’t been because of idleness. Quite the opposite.
Most of my time this year has been dedicated to gaining a strong understanding of deep learning. I built a solid theoretical foundation by reading Ian Goodfellow’s wonderful book Deep Learning and watching lecture videos from Stanford’s CS231n, Convolutional Neural Networks for Visual Recognition. I rounded out my knowledge by reading various papers, watching talks on YouTube, etc.
January 29, 2016
This past holiday season, I once again spent my “time off” working on a solution to a Kaggle holiday challenge. This year, Santa needed help after his magic sleigh was stolen!
Instead of delivering all the presents in one trip, Santa was forced to make multiple trips using his non-magical sleigh. The sleigh had a weight limit and the poor reindeer were in danger of exhaustion from all the trips. Santa needed a plan for delivering the presents that minimized reindeer weariness.
November 3, 2015
For the Whenology project, I needed to visually describe key concepts related to leaf color change during Fall. The obvious answer was to simply draw some trees! But how?
I knew that D3js could import SVG files, so the first step was to draw a tree. I started with a public domain picture of a leaf and used Adobe Illustrator’s Image Trace feature to turn it into an SVG outline.
October 2, 2015
I like to solve hard problems. So much so that when I get a particularly good one it’s often difficult to do anything else. This personality quirk has been great for my engineering career but not so good for consistently blogging. Maybe you’ve noticed.
The good news is that I’m finally coming up for air and have lots to blog about.
Since February, I’ve been collaborating with scientists from Acadia National Park, the Earthwatch Institute, and the Schoodic Institute to study how climate change may be altering the life-cycles of and interactions between species at the park.
January 29, 2015
In parts 1 and 2 , I described how I used the AMS and HyperLogLog online algorithms to estimate the frequency distribution of values in multiple integer sequences. I was doing this for a friend that needed a compute and space efficient way to analyze thousands of such streams.
Although the solution I came up with in Part 2 worked, it didn’t express the “skew” in human understandable terms like “20% of the values accounted for 80% of the occurrences”.