The Blog


GerryCamp Week 4

Week 4 is done! We’ve only got two weeks left in the main program and things are starting to come together in a way that’s really, really exciting. This week I worked with the group building the Markov chain Monte Carlo (MCMC) simulator. Briefly, an MCMC sampler solves the problem of ‘we know how to draw from distribution A but...
30 Jun 2018

GerryCamp Week 3

Week three is done (and I guess week four has started, but I’m slow…) and I worked on All Things Spectral. In short, to any districting plan, either as a collection of shapes or an abstract graph, we can associate a certain operator called the Laplacian which measures how geometrically similar the neighborhood of each point is to that of...
25 Jun 2018

GerryCamp Week 2

Okay, the second week of redistricting summer camp is over and a lot has happened. Our first week of independent project work has highlighted many of the challenges we will face as researchers over the remaining weeks as well as demonstrated proof-of-concept for several promising new areas to look at. The research approach is very divide-and-conquer, so we all split...
18 Jun 2018

GerryCamp Week 1

This summer I am working with the MGGG as a Graduate Fellow with the Voting Rights Data Institute, a.k.a. gerrymandering summer camp a.k.a. GerryCamp. I’m writing these weekly posts both as a way of documenting my work and experience as well as a sort of proof-of-life for my friends and colleagues I (temporarily) abandoned in Philadelphia. This week mostly consisted...
09 Jun 2018

Strategic Classification from Revealed Preferences

Our paper Strategic classification from revealed preferences was just accepted at EC ‘18! Not only is it exciting that our work is being published, but this is also my first publication which, I think, officially makes me an academic. At the very least, my Erdős number is 4. Our paper examines a setting in which self-interested agents interact with a...
23 Apr 2018

An Incorrect Conjecture and a Theorem From Graph Theory

Last week, I casually asked a few of my colleagues about a little graph theory conjecture I had. It took me about five minutes to disprove it when I got back to my desk, but interestingly enough something similar to the converse of the statement is true (and a stronger version of it, to boot!). Two of my favorite things...
10 Dec 2017

Gerrymandr - Day 3

This week, I am at the Metric Geometry and Gerrymandering Group’s conference/workshop in Somerville, and, given my poor level of information retention from EC earlier this summer, I wanted to take a more deliberate approach to absorbing and processing the information that’s being given to me this week. I’m taking notes during the talks and I’ve decided to write these...
10 Aug 2017

Gerrymandr - Day 2

This week, I am at the Metric Geometry and Gerrymandering Group’s conference/workshop in Somerville, and, given my poor level of information retention from EC earlier this summer, I wanted to take a more deliberate approach to absorbing and processing the information that’s being given to me this week. I’m taking notes during the talks and I’ve decided to write these...
09 Aug 2017

Gerrymandr - Day 1

This week, I am at the Metric Geometry and Gerrymandering Group’s conference/workshop in Somerville, and, given my poor level of information retention from EC earlier this summer, I wanted to take a more deliberate approach to absorbing and processing the information that’s being given to me this week. I’m taking notes during the talks and I’ve decided to write these...
08 Aug 2017

Graphs are PAC-Learnable from Subgraphs

This semester I took a course on Computational Learning Theory, which deals with the statistical and computational underpinnings of machine learning. As part of our final project, Hadi Elzayn and I proved that graphs are Probabily-Approximately Correct (PAC)-learnable from labeled supergraph/subgraph examples in polynomial sample complexity. Here’s a presentation of that proof as well as a fun little way of...
25 May 2017

First Year, First Post

So I’m one academic year into grad school. A lot has happened in the year since I graduated from Colby: I moved from Waterville to Philadelphia, I completed two semeseters of graduate level CS coursework with minimal undergrad training in the discipline, I’ve started working on some interesting research problems, and I’ve met some interesting people. I’ve now tried three...
24 May 2017