Optimal Transport for Machine Learning Reading List

Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026

Optimal transport provides geometry-aware distances between distributions. Theory, computation, and applications to ML.

Optimal Transport for Machine Learning: 10 key papers

1. Computational Optimal Transport
   Peyré and Cuturi. arXiv 2018.
2. Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distances
   Cuturi. arXiv 2013.
3. Wasserstein GAN
   Arjovsky et al. arXiv 2017.
4. Sliced Wasserstein Distance for Learning Gaussian Mixture Models
   Kolouri et al. arXiv 2017.
5. Data-driven Distributionally Robust Optimization Using the Wasserstein Metric: Performance Guarantees and Tractable Reformulations
   Esfahani and Kuhn. arXiv 2015.
6. Optimal Transport for Domain Adaptation
   Courty et al. arXiv 2015.
7. Approximating 1-Wasserstein Distance with Trees
   Yamada et al. arXiv 2022.
8. Improved Training of Wasserstein GANs
   Gulrajani et al. arXiv 2017.
9. Large-Scale Optimal Transport and Mapping Estimation
   Seguy et al. arXiv 2017.
10. Stochastic Optimization for Large-scale Optimal Transport
    Genevay et al. arXiv 2016.

Related reading lists

• Flow Matching & Schrödinger Bridges
• Robust & Risk-Sensitive RL
• Quality-Diversity & Novelty Search
• Score-Based Generative Modeling
• Optimal Control & LQR/LQG
• Domain Adaptation & Domain Generalization
• Linear MDPs & Linear Function Approximation
• Adversarial Reinforcement Learning & Robustness to Adversaries