Changing model architecture
165 views
Skip to first unread message
Nate Riek
Apr 21, 2023, 4:05:26 PM4/21/23
to physionet-challenges
For the unofficial phase I have tried some different models. I have some new ideas that are very different than what I have submitted for the unofficial phase. Will this be a problem for my abstract if my plan during the official phase is very different from the model I have a score for?
Most abstracts I have written are about what I have already done. Should I focus on what I have already done or my plan for future models?
Thanks,
Nate
Nate
PhysioNet Challenge
Apr 21, 2023, 4:10:06 PM4/21/23
to physionet-challenges
Dear Nate,
Please do describe what you have done and what you plan to do, and please be clear about which is which, i.e., what you have already done and are reporting scores for, and what you plan to do. We expect that your approach will evolve over the course of the Challenge, and we expect that the approaches presented at the CinC conference will have evolved from the approaches described in the CinC abstract submissions. (Yes, you can update your abstract for your CinC proceedings paper.)
The goal is to convince the CinC abstract reviewers that you will do interesting work during the official phase, and that we need to hear about your work at CinC. For example, if you write that you tried several basic machine learning models during the unofficial phase, and you'll try several deep learning models during the official phase, but you don't share much more than that, then they may be skeptical. If you talk about the specifics of your current models, and how you plan to preserve its strengths and address its deficiencies with new models, then they should be more receptive. Please use complete sentences, be specific, and include your scores from the leaderboard:
Please do describe what you have done and what you plan to do, and please be clear about which is which, i.e., what you have already done and are reporting scores for, and what you plan to do. We expect that your approach will evolve over the course of the Challenge, and we expect that the approaches presented at the CinC conference will have evolved from the approaches described in the CinC abstract submissions. (Yes, you can update your abstract for your CinC proceedings paper.)
The goal is to convince the CinC abstract reviewers that you will do interesting work during the official phase, and that we need to hear about your work at CinC. For example, if you write that you tried several basic machine learning models during the unofficial phase, and you'll try several deep learning models during the official phase, but you don't share much more than that, then they may be skeptical. If you talk about the specifics of your current models, and how you plan to preserve its strengths and address its deficiencies with new models, then they should be more receptive. Please use complete sentences, be specific, and include your scores from the leaderboard:
Best,
Matt
(On behalf of the Challenge team.)
Please post questions and comments in the forum. However, if your question reveals information about your entry, then please email info at physionetchallenge.org. We may post parts of our reply publicly if we feel that all Challengers should benefit from it. We will not answer emails about the Challenge to any other address. This email is maintained by a group. Please do not email us individually.
Matt
(On behalf of the Challenge team.)
Please post questions and comments in the forum. However, if your question reveals information about your entry, then please email info at physionetchallenge.org. We may post parts of our reply publicly if we feel that all Challengers should benefit from it. We will not answer emails about the Challenge to any other address. This email is maintained by a group. Please do not email us individually.
Reply all
Reply to author
Forward
0 new messages