Call for Contributions
We solicit submission of original research at the interface between optimal transport theory, statistics, optimization, machine learning and applications.
Authors can submit work that overlaps with previously published or submitted work, as long as it adds a new perspective on that work.
Selected submissions will be presented in spotlight talks.
Note that the workshop will not have proceedings and any work that has been or
will be submitted to a reviewed Machine learning conference is not considered double
submission.
Important Dates
- Submission deadline: Friday, 17 September 2021
- Late submission deadline, 2 October 2021
- Author notification: Friday, 15 October 2021
Submission Website
All submissions must be processed using the OpenReview website.
The submissions should be anonymized following NeurIPS policy.
We guarantee that all works submitted before September 17, 2021 (23:59 PDT)
will be given full consideration and will be reviewed by at least three
reviewers. Late submissions until October 2nd will be considered but might have fewer reviews.
Note that at least one of the co-authors should accept to perform reviews for the OTML Workshop.
Scope and Topics
Topics of interest to the workshop broadly include
Estimation of Optimal Transport Couplings and Maps
- Resolution of the Kantorovich problem and its variants (semi-discrete, continuous) at large scales
- Estimation of Monge maps between measures using parameterized or constrained families of potentials
- Estimation of Monge maps across time using JKO formulations.
Generalizations of Optimal Transport
- Unbalanced formulations: optimal transport between measures of different total mass
- Quadratic formulations of the optimal transport problem, Gromov-Wasserstein metrics. optimal transport with rigid transformations.
- Multi-marginal formulations for optimal transport
- Quantum optimal transport
- Martingale optimal transport and time constraints in the estimation of optimal transport
Optimal Transport as a Learning Methodology
- Optimal transport costs as a loss: GAN type results, minimization of Wasserstein distances between parameterized measures and data measures.
- Optimal transport to define data transformations: domain adaptation, clustering, incorporation into MCMC methods.
- Learning of ground metrics and costs, optimal transport carried out on projections (sliced and higher dimensional)
Statistics and Theory
- Interplay between optimal transport and statistics: properties of finite sample optimal transport estimators between continuous measures, either seen as Monge maps or couplings.
- Optimal transport as a criterion to study the convergence of sampling methods.
- Optimal transport as a tool to define convexity, sorting and quantiles
- Study of the complexity of optimal transport algorithms such as the Sinkhorn iteration and variants.
Optimal Transport in Applications
- High-dimensional applications: Natural Language / Word Embeddings, Biology, Vision, etc.
- Low-dimensional applications: Graphics, shapes, imaging, univariate measures, etc.
Style
Authors should use the standard NeurIPS 2021 style files.
The recommended length for submissions is between 6 and 9 pages (main paper
without references). Note that we will not accept supplementary material but an
annex/supplementary can be added a the end of the paper and will not be counted in the pages.
All submissions must be processed using the OpenReview website.
The submissions should be anonymized following NeurIPS policy.
We guarantee that all works submitted before September 17, 2021 (23:59 PDT)
will be given full consideration and will be reviewed by at least three
reviewers. Late submissions will be considered as well until October 2nd but
might have fewer reviews.