Research map / Learning theory and algorithms
Optimal transport: research map
462 papers on Optimal transport in Learning theory and algorithms, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 7 approaches. The busiest year so far is 2026.
What it is
Optimal transport measures how much "work" it takes to move one probability distribution onto another, giving the Wasserstein distance. It is used to compare distributions, align datasets, train generative models and build bridges between distributions.
Terms you will see on this page:
- Optimal transport (OT)
- the cheapest way to move mass from one distribution to another.
- Wasserstein distance
- the cost of that optimal move.
- Sinkhorn algorithm
- a fast iterative solver for OT with entropy regularisation.
Start reading with the most cited papers here:
- Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration (NeurIPS 2017 · 292 citations)
- Neural Architecture Search with Bayesian Optimisation and Optimal Transport (NeurIPS 2018 · 262 citations)
Within Learning theory and algorithms, its share grew from 3.8% in 2023–24 to 5.8% in 2025–26 (84 → 142 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Optimal transport in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Sliced Wasserstein variants rising: 8% → 18% of the topic (7 → 26 papers)
- Neural optimal transport rising: 23% → 30% of the topic (19 → 42 papers)
- Sinkhorn and entropic OT algorithms rising: 18% → 21% of the topic (15 → 30 papers)
- Transport map estimation falling: 35% → 15% of the topic (29 → 22 papers)
Papers from 2025–2026 that are already cited most:
- Schrödinger Bridge Matching for Tree-Structured Costs and Entropic Wasserstein Barycentres (NeurIPS 2025 · 0 citations)
- Tree-Sliced Wasserstein Distance: A Geometric Perspective (ICML 2025 · 0 citations)
- Towards Understanding Gradient Dynamics of the Sliced-Wasserstein Distance via Critical Point Analysis (ICML 2025 · 0 citations)
Approaches and key papers
Optimal transport methods · 311 papers
Keywords: optimal transport, transport ot, sinkhorn
Approaches in this cluster:
- Neural optimal transport (131 papers)
Learn transport maps and distributions with neural, generative and loss-based formulations. - Transport map estimation (88 papers)
Estimate and regularize transport maps with statistical and low-rank guarantees. - Sinkhorn and entropic OT algorithms (81 papers)
Analyze fast solvers for entropic and unbalanced optimal transport. - Schrodinger bridge matching (11 papers)
Solve entropic OT through Schrodinger bridges with iterative fitting and benchmarks.
Most cited and most cited since 2024:
- Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration (NeurIPS 2017 · 292 citations)
- Neural Architecture Search with Bayesian Optimisation and Optimal Transport (NeurIPS 2018 · 262 citations)
- Hierarchical Multi-Marginal Optimal Transport for Network Alignment (AAAI 2024 · 15 citations)
- An Optimal Transport View for Subspace Clustering and Spectral Clustering (AAAI 2024 · 6 citations)
Wasserstein distances · 151 papers
Keywords: wasserstein distance, sliced, sw
Approaches in this cluster:
- Wasserstein barycenters and GW (71 papers)
Compute Wasserstein barycenters, projection-robust and Gromov-Wasserstein distances with new optimization methods. - Sliced Wasserstein variants (55 papers)
Develop sliced and tree-sliced Wasserstein distances with faster approximations and kernels. - Wasserstein losses for generative models (25 papers)
Train generative models with Wasserstein and sliced losses, with asymptotic and convergence guarantees.
Most cited and most cited since 2024:
- Gromov-Wasserstein Averaging of Kernel and Distance Matrices (ICML 2016 · 216 citations)
- Sliced Wasserstein Kernel for Persistence Diagrams (ICML 2017 · 151 citations)
- A New Robust Partial p-Wasserstein-Based Metric for Comparing Distributions (ICML 2024 · 3 citations)
- Semidefinite Relaxations of the Gromov-Wasserstein Distance (NeurIPS 2024 · 3 citations)
Related topics in Learning theory and algorithms
- Matrix and tensor methods (1,467)
- Sample complexity (1,553)
- Combinatorial and search optimization (1,576)
- Fairness (550)
- Submodular and game algorithms (1,064)
- Prediction and decision losses (2,400)
- Kernels and regression theory (1,272)
- Clustering (998)
