Skip to main content Link Menu Expand (external link) Copy Copied
Splash photo of MBZUAI
Third Workshop on Seeking Low‑Dimensionality in Deep Neural Networks
January 2023, MBZUAI

Key Dates

  • Workshop dates: 3rd - 7th January, 2023; in person at W Hotel, Abu Dhabi
  • Submission portal opens: October 1st, 2022
  • Paper submission deadline: November 6th, 2022
  • Acceptance and travel grant notification: November 20th, 2022

How to Submit Your Work

We will use OpenReview to manage submissions. Submit your work here:

See the submission guidelines and topics of interest.

Logistics for Accepted Papers

Accepted works will be expected to present a poster describing the work in-person at the workshop.

Travel grants are available to support authors of accepted papers: see the travel page for details.

A small subset of the top accepted papers will be recommended for inclusion in a future special issue of the IEEE Journal of Selected Topics in Signal Processing.

Topics of Interest

Topics of interest include, but are not limited to, connections between low-dimensional models and the theory, architectures, algorithms, and applications of deep neural networks:

  • Theory: approximation, generalization, robustness, compact and structured representations
  • Optimization: Benign non-convex optimization, implicit bias analysis, convergence guarantees
  • Architectures: compact/model-based/neuro-inspired/invariant neural networks
  • Algorithms: pruning, sparse training, robust training
  • Applications: generative models, resource/data-efficient learning, inverse problems

Submission Guidelines

We aim to showcase:

  • The latest research innovations at all stages of the research process, from work-in-progress to recently published papers
  • Position or survey papers on any topics relevant to this workshop (see the call above)

Concretely, we ask members of the community to submit a conference-style paper (from four to eight pages, with extra pages for references) describing the work. Please also upload a short (250 word) abstract to OpenReview. Do not anonymize submissions.

Papers should be written using the NeurIPS 2022 style files, available here.

OpenReview submissions may also include any of the following supplemental materials that describe the work in further detail.

  • A poster (in PDF form) presenting results of work-in-progress.
  • A link to a blog post (e.g.,, Medium) describing results.
  • Appendices with detailed derivations and additional experiments.

This workshop is non-archival, and it will not have proceedings. We permit under-review or concurrent submissions.

Reviewing will be performed in a single-blind fashion (authors should not anonymize their submissions).