mlsanityData to 14 Oct 2026

Fitted value iteration methods for bicausal optimal transport

arXiv 2306.12658 ↗First posted 22 Jun 2023stat.ML cs.LG q-fin.MFcited by 2

Not on the map. Why?

Abstract (arXiv)

We develop a fitted value iteration (FVI) method to compute bicausal optimal transport (OT) where couplings have an adapted structure. Based on the dynamic programming formulation, FVI adopts a function class to approximate the value functions in bicausal OT. Under the concentrability condition and approximate completeness assumption, we prove the sample complexity using (local) Rademacher complexity. Furthermore, we demonstrate that multilayer neural networks with appropriate structures satisfy the crucial assumptions required in sample complexity proofs. Numerical experiments reveal that FVI outperforms linear programming and adapted Sinkhorn methods in scalability as the time horizon increases, while still maintaining acceptable accuracy.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 27 of its 61 references that are in the map. First posted on arXiv 22 Jun 2023.

Builds on

27 of 61 references.

References · 27

  1. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  2. 2018Computational Optimal Transportcited by 780
  3. 2017Nonparametric regression using deep neural networks with ReLU activation functioncited by 433
  4. 2019Information-Theoretic Considerations in Batch Reinforcement Learningcited by 287
  5. 2017Near-linear time approximation algorithms for optimal transport via Sinkhorn iterationcited by 223
  6. 2019A Theoretical Analysis of Deep Q-Learningcited by 215
  7. 2016Stochastic Optimization for Large-scale Optimal Transportcited by 189
  8. 2018Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: optimal rate and curse of dimensionalitycited by 162
  9. 2017Large-Scale Optimal Transport and Mapping Estimationcited by 123
  10. 2021Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmarkcited by 96
  11. 2020COT-GAN: Generating Sequential Data via Causal Optimal Transportcited by 71
  12. 2021Risk Bounds and Rademacher Complexity in Batch Reinforcement Learningcited by 42
15 more
  1. 2019On the Efficiency of Entropic Regularized Algorithms for Optimal Transportcited by 35
  2. 2021Quantitative Stability of Regularized Optimal Transport and Convergence of Sinkhorn's Algorithmcited by 22
  3. 2016Causal transport in discrete time and applicationscited by 21
  4. 2020Approximating smooth functions by deep neural networks with sigmoid activation functioncited by 18
  5. 2022Computational methods for adapted optimal transportcited by 12
  6. 2022Convergence Rates for Regularized Optimal Transport via Quantizationcited by 12
  7. 2016Causal optimal transport and its links to enlargement of filtrations and continuous-time stochastic optimizationcited by 11
  8. 2015Local Rademacher Complexity Bounds based on Covering Numberscited by 10
  9. 2018Extended Mean Field Control Problems: stochastic maximum principle and transport perspectivecited by 7
  10. 2018Computation of optimal transport and related hedging problems via penalization and neural networkscited by 7
  11. 2017Marginal and dependence uncertainty: bounds, optimal transport, and sharpnesscited by 4
  12. 2022Stability and Sample Complexity of Divergence Regularized Optimal Transportcited by 4
  13. 2020Cournot-Nash equilibrium and optimal transport in a dynamic settingcited by 3
  14. 2023Distributionally robust Kalman filtering with volatility uncertaintycited by 2
  15. 2018Transport plans with domain constraintscited by 1

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 1.

2 papers.

  1. 2024Bisimulation Metrics are Optimal Transport Distances, and Can be Computed Efficientlycited by 6
  2. 2025Nested Optimal Transport Distancescited by 0