Skip to main content

C

o

e

n

D

.

N

e

e

d

e

l

l

My work was recently featured in WIRED thanks to a discovery about art by my former colleagues Trent Davis and Wilma Bainbridge! Check it out here!

I am a Pre-Doctoral Researcher at the Computational Social Science Lab at University of Pennsylvania. I previously worked at Microsoft Research, NYC in the Computational Social Science group. I have a Masters of Computational Social Science from the University of Chicago with my thesis research on using deep learning to estimate the memorability of images. As an undergraduate at Washington University in St. Louis, I majored in Physics and Economics, and minored in the Philosophy of Science. My work focuses on developing novel methods for social scientific concepts. My hobbies include analyzing video games, cooking, and playing classical guitar.

This site is intended to be a place for me to post my projects, some personal, some academic. They are mostly computational in nature, but some are philosophical.

Featured Module: Where is The Cloud?

A small demo showing how to use a dataset to generate a three.js model. This was done using a dataset of cloud data centers retrieved from Cloud Infrastructure Map. Code is availible on GitHub. Different providers are shown in different colors. Alibaba Cloud is shown in White. Amazon Web Services is shown in Orange. Google Cloud is shown in Pink. Huawei Cloud is shown in Red. IBM Cloud is shown in Dark Blue. Microsoft Azure is shown in Green. Oracle Cloud is shown in Purple. Tencent Cloud is shown in Light Blue. Locations are not exact, and a small jitter is applied to prevent overlaps. Globe by Denys Almaral.

Ongoing Projects

Data Viz in Hugo

Neovim for Data Science

News Observatory

Featured Posts

Data Visualizations: Overview

Problem: pretty, dynamic data visualization frameworks are either powerful, but difficult to use (like D3.js) or limited in scope and comparatively heavy for the web (like observable). Using Hugo modules and shortcodes, we can create a system of modules for integrating data visualizations with markdown blogging. This post introduces core concepts, and details the development of a simple visualization.

https://coen.needell.org/post/lj/barplot/ Coen D. Needell

Neovim for Data Science, Part 1: Scripting

Vim, “the ubiquitous text editor”, has been with us since the 1980s, with new programmers discovering its arcane power every day. It was originally developed for the Atari ST under the name “Stevie” (ST Editor for VI Enthusiasts), and was later ported to Unix and OS/2 (a precursor to Windows). Originally, vim was simply an Atari port of vi1:, which in turn was the visual mode for the command line text editor ex2.

https://coen.needell.org/post/neovim_for_data_science_1/ Coen D. Needell

Comparing ResMem and MemNet

In my previous posts about Memorability (see the project link above), I’ve been talking about the performance of my models fairly matter-of-factly. I’ve been comparing their scores on things, reporting them in abstracts, and talking about how one model performs better than another, and why I think that is happening. Some questions arise though, for example, why did I get such a vastly different score with MemNet than what Khosla et al.

https://coen.needell.org/post/memnetch4/ Coen D. Needell

Collaborative Filtering and Innovation: A Case Study from Steam Recommendations

Introduction The modern world is flooded with options; which TV show to watch, which video games to play, which espresso machine to buy. E-commerce is characterized by its quantity. This doesn’t, however, imply that e-commerce platforms are inundated with low quality goods. Truly fantastic media is released every year. 2019 saw the release of a game Disco Elysium which has been lauded as one of the best CRPGs (Computer Role Playing Games) of all time, even though the so-called golden age of the genre is long past(Metacritic n.

https://coen.needell.org/post/steam_algorithms/ Coen D. Needell

ResMem and M3M

In my last post on computer vision and memorability, I looked at an already existing model and started experimenting with variations on that architecture. The most successful attempts were those that use Residual Neural Networks. These are a type of deep neural network built to mimic specific visual structures in the brain. ResMem, one of the new models, uses a variation on ResNet in its architecture to leverage that optical identification power towards memorability estimation.

https://coen.needell.org/post/memnetch2/ Coen D. Needell

ResMem Release

A user-ready version of ResMem is now available on PyPI! The model included in the package is designed to estimate the memorability of an input image but is not intended for feature space analysis. The model is optimized for accuracy by allowing the ResNet features to retrain. The model included in the resmem package has been dubbed “ResMemRetrain” for this reason. Statistically, the retrained model performs better than the model where ResNet is as-is, receiving a Spearman rank correlation of 0.

https://coen.needell.org/post/resmemrelease/ Coen D. Needell

MemNet: Models for Predicting Image Memorability

Memnet1 was an attempt to build a neural network-based model to predict the memorability of an image. This attempt was carried out by Khosla et al. at the Computer Science and Artificial Intelligence Labs at MIT to moderate success. It is the most commonly used neural network regression for this purpose, and has been used and cited in many research papers since publication. There are some problems, however. Memnet was built in Caffe, a deep learning framework which has been defunct since shortly after Memnet’s publication.

https://coen.needell.org/post/memnetch1/ Coen D. Needell

Blaseball Ticker Analysis

Introduction In the last week, large swathes of the Internet have been enamored with the simulated sport of blaseball. Blaseball describes itself as “baseball at your mercy, baseball perfected”. I’ve been describing it as “Nightvale-esque simulated baseball with rules decided by a voting system”. It’s fun, that’s for sure. Blaseball saw a boom after being picked up by a number of news outlets, and a number of blaseball users thought “now, this is fun, but I want to know it like I know myself”.

https://coen.needell.org/post/blaseball_ticker_analysis/ Coen D. Needell

Accelerated Gammatones

Introduction Last winter, I worked on a personal project I call ongaku (from the Japanese for ‘music’). This was an attempt to use manifold learning to create a metric space for music. The preprocessing relied heavily on a method called (Valero and Alias 2012). This method was intended to replace Mel Frequency Cepstral Coefficients. Where Mel Frequency is a logarithmic transformation of sound frequency, in an attempt to simulate human perception of sound.

https://coen.needell.org/post/gammatones/ Coen D. Needell

Sex, Conspiracy, and Video Games: Urban Legends on the Internet

Introduction Urban Legends, or Contemporary Legends, are a type of folklore popularized by Jan Harold Brunvand in his book: The Vanishing Hitchhiker: American Urban Legends & Their Meanings. Urban Legends are considered to be the modern continuation of the human tradition of folklore and legends. Folklore does not occur exclusively in so-called primitive or traditional societies, and by studying modern folklore in the same way that we study the folklore of traditional societies, we can also learn about modern societies(Brunvand 2003).

https://coen.needell.org/post/urban_legends/ Coen D. Needell