Initial Perspectives on Research

The goal of the Distributed REsearch Apprenticeships for Master’s (DREAM) experience is to provide students with research experience to help them better understand what goes into research, what doing a phD would be like, and how choices around whether or not to do research might affect their options moving forward. The program also aims to let students gain specific research skills that they could apply in their own research endeavors.

How do I see research now? Currently, the way I see research is based on my previous experience. In undergrad, I studied in a liberal arts program that had a broad array of foundation courses. The goal in these courses was not to teach us facts about a particular discipline, but to teach us how to think like a professional or researcher in a given discipline. While I never conducted formal research in undergrad, I did learn about qualitative and quantitative research, literature reviews, the need for ethics committees, to always think critically about specific methods and claims in published research, and not to take something as true just because it is published. I learned that a statistician can make a data set look a lot of different ways based on how they decide to approach is, and researchers should always check for and identify personal bias.

In my Master’s program, I’ve had the privilege to work on a few research projects. One was algorithm research on q-learning for optimizing optical networks, and one was creating a regression to model and predict covid-19 cases. In these cases, I learned that computer science research relies heavily on getting up to speed on current best practices in the given arena, and then trying or adding something new. In the first project, we were working with a professor who did have domain knowledge he was able to share, while in the second paper the research did not really seem usable, because we were far from experts in machine learning and covid-19. The second paper was more of a ML learning experience than a helpful research contribution.

What do I hope to get out of this position?

In this position, my hopes for learning outcomes are:

1. To be able to articulate what research looks like in the arena of data visualization
2. Learn some new technical skills, and gain exposure to data viz tools
3. Learn how a dataset is curated and develop confidence in working with datasets
4. Learn about how the agile process can be applied to research

Where are my biggest hesitancies?

My biggest hesitancies in this project are learning new tools and a new language. While I do enjoy expanding my toolbelt and learning more about technology, the learning curve that comes with adopting a new tool is one of the most challenging parts of computer science for me.

I do like challenges, working past limiting beliefs, and expanding my potential, and working through the discomfort of learning new technological tools is just part of that. While this is my biggest hesitancy, I also know this program is a supportive container to try new things, and lean into the skills I am less confident with.

Where are my biggest excitements?

My biggest excitements are learning more about irresponsible cases of AI, playing with new tools, expanding my community, and getting to write a bit.

I’ve taken a machine learning course in my Master’s program, and am signed up for an AI course this fall. Coming from a humanities background, I am very interested in the context that AI exists in, and how it impacts the world at large. My computer science courses have been mainly technical, so I’m excited to be in a project where the content will give me some more exposure to the human stories around AI.

While learning new tools is a biggest hesitancy, I am excited for the rewards that come with getting to play in new ways once I am more comfortable with the tools.

Lastly, I’m excited to make new friends and colleagues in this field, learn from what they have to share, and then hopefully write and reflect on the pieces I learn along the way.

Project scope and initial plan

Our research project is on data visualization for irresponsible artificial intelligence data. This means that we research and improve the best practices for visualizing this kind of data, with domain experts being our main audience. We are working in observable notebook with d3 and Vega-Lite.

We have a specific dataset we are working with, that has been curated from various articles about irresponsible uses of AI.