[edit]
Data-centric Machine Learning Research
The Journal of Data-centric Machine Learning Research (DMLR) is a new member of the JMLR family, aiming to provide a top archival venue for high-quality scholarly articles focused on the data aspect of machine learning research. DMLR aims to maintain the high-quality bar of JMLR with a rigorous review process and a high-profile, dedicated editorial board and review board with diverse expertise and representation from all adjacent fields. We will continue to refine and tailor our process to better serve the special nature of data-centric ML research. We also hope that DMLR will become the venue for innovations that are tailored to the special nature of data-centric ML research.
For more information on the DMLR journal, visit our blog here.- 2024.07.27: We will share the latest DMLR Journal updates with you during the announcements of the DMLR Workshop at ICML. Don't miss it if you are around!
- 2024.06.01: Check out the accepted papers so far!
- 2023.10.11: DMLR Journal is accepting submissions; check out the Submissions page for guidelines and upload your paper on OpenReview!
- Newsha Ardalani (Meta)
- Isabelle Guyon (Google)
- Neil Lawrence (University of Cambridge)
- Joaquin Vanschoren (TU Eindhoven)
- Ce Zhang (University of Chicago)
- Merve Gürel (TU Delft)
The DMLR employs a single-blind review system in which authors' identities are not anonymized. This is because data work often includes the hosting of data, which usually cannot be anonymized. Authors are required to create author profiles on OpenReview and disclose any potential conflicts of interest when submitting their papers. Additionally, they must provide information concerning human subjects reporting (IRB), funding, competing interests, and any conflicts not covered in their institutional history.
The review and publication procedures of DMLR closely resemble those of JMLR and TMLR
Briefly, upon submission, an Editor-in-Chief (EIC) will assign an Action Editor (AE) to assess whether the submission aligns
with the Acceptance Criteria, including scope, completeness, and compliance with the DMLR standards. In certain cases,
the EIC may reject the submission outright before the AE assignment. If the AE opts to proceed with the review,
three reviewers will be recruited to evaluate and provide feedback on the submission. Subsequently, an open-ended phase
for rebuttals, discussions, and revisions will enable authors to engage with reviewers and refine their paper. Lastly, the
AE will make a decision, which may be to accept the submission as is, accept it with minor revisions, or reject it
(with an option to make revisions and resubmit it anew). Submissions will be public on OpenReview only when they are accepted. Before submitting to DMLR, we recommend that you carefully
read the Submission page as well as Reviewer Guidelines and Action Editor Guidelines, which details DMLR's scope, criteria, and review process.
DMLR is published electronically, and the International Standard Serial Number (ISSN) is currently pending.
Back to the top