Next week will be all about scale! I will take our base code and make it work with many different models and datasets. More on that next week.
Thanks for clicking through despite this week’s delay! The weekly publishing schedule of these updates will now shift slightly backward.
I do this because the alignment with my supervisor and co-author is usually on Tuesdays, so I will publish these updates after the alignment meeting, and not anymore around Sundays.
This week, I (remotely) visited the 3rd ACM International Conference on AI in Finance. The conference fits my research area, and it gave me a sense of the type of articles that are in demand and how these papers are presented.
You can follow these updates: Substack Blog Telegram WhatsApp LinkedIn Medium Twitter Calendly
What Happened Since Last Week?
We set up Weights & Biases to track our model during training, and we created a standardized, automated spreadsheet output for our model results. This spreadsheet will likely become the basis of our first article.
While I was listening to conference presentations for three days straight, my co-author did further testing of our code on the DGX-2.
What Were the Biggest Obstacles?
I enjoyed listening to the conference, but of course, it reduced the time I could work on coding and writing.
Which Goals Did I Meet?
Participate in the 3rd ACM ICAIF '22 conference.
Decide on whether to prepare a dataset ourselves or take an existing dataset. We will prepare our dataset! The decision was made.
Which Goals Did I Miss?
Expand the experiments to additional models. (Although credit goes to my co-author for already adding one more model. We currently have two working models.)
Was It a Good Week?
Yes. As mentioned last week, we had a very abrupt change in plans and spontaneously planned another article. Until yesterday, I have not had the chance to align on that with my supervisor. But he confirmed yesterday that this is a good path, and I’m now very excited about it. I think there might be a chance that we can have an article about our small new idea ready by the end of the year, which would be a great achievement.
Short-Term Tasks for The Coming Week
Next week will be all about scale because I will take the existing code and make it work with arbitrary question answering models and arbitrary question answering datasets:
Expand the experiments to additional models. (Even more models. We currently only have two. I will scale this to 10 models.)
Expand the experiments to additional datasets.
About “75-Step Journey Toward a Ph.D. in Natural Language Processing”
You will, from now on, witness my grind. Feel my blood, sweat, and tears.
With this series of articles, you become a real-life weekly witness of my dissertation progress, all in 75 steps. This has multiple purposes:
1) Forcing myself to keep moving through the power of public shame!
2) Helping other (prospective) Ph.D. students to stay motivated and to show that hard times are normal when going through this process.
3) Getting support from the community when I go through hard times.
Share this with your Ph.D. student friends: Substack Blog Telegram WhatsApp LinkedIn Medium Twitter Calendly.
Read More From the 75 Steps Toward a Ph.D. in NLP Series
2022-08-20: Update 1/75 - Kicking Off the Journey Toward a Ph.D. in NLP
2022-08-28: Update 2/75 - Literature Review
2022-09-04: Update 3/75 - Back on Track and Back to Vallendar
2022-09-10: Update 4/75 - Long Test Runtime; Retriever Works
2022-09-18: Update 5/75 - Jour Fixe Joy
2022-09-26: Update 6/75 - Reading Group
2022-10-02: Update 7/75 - Leaving the Phone at Home
2022-10-09: Update 8/75 - Finding a Conference
2022-10-16: Update 9/75 - Dataset - Make or Take
2022-10-23: Update 10/75 - Still Unsure About the Dataset
2022-10-30: Update 11/75 - NVIDIA DGX-2 and Swiss Cheese
2022-11-10: Update 12/75 - Three Days of Conference via Zoom