Question Solved Locked

How to measure the statistical significance of A/B test results in ML experiments? (76)

Back to Forum
Growing 140 pts 0 followers
Canara College · Posted
Astronomy & Astrophysics

Context

I am preparing to submit my first research paper to a Q1 journal (IEEE Transactions on Neural Networks). This will be my first submission and I am unsure about a few things.

Questions:

  1. Is it mandatory to submit the code and dataset along with the paper, or is that only required upon acceptance?
  2. My results show a 2.3% improvement over the SOTA baseline. Is that considered significant enough?
  3. How long does the review process typically take for IEEE Transactions journals?

I have read the author guidelines but they are somewhat ambiguous on point 1. Would appreciate guidance from anyone who has published in similar venues.

Sign in to interact

Sign in to join the discussion.

3 Replies

Oldest Top
0
Farhan Mirza · Accepted answer

I think you are conflating two separate problems here. The publish-or-perish culture creates pressure, yes.but the solution is not to criticize researchers who are navigating an unfair system. The real problem is with evaluation committees who treat publication count as a proxy for research quality. Fix the evaluation, and the incentives change.

Also worth noting: open review has been implemented in some fields and has its own problems. It can disadvantage early-career researchers who fear retaliation from senior colleagues they have reviewed critically.

0
Preethi Anand Distinguished 690 pts · Accepted answer

Check whether your institution has a Springer/Elsevier waiver agreement.many NITs and IITs do.

0
Arjun Venkatesan Growing 195 pts · Accepted answer

Overleaf + GitHub integration is underrated. You get version control and collaboration in one place.