Convex relaxation approaches for high-dimensional optimal transport
By: Yuehaw Khoo, Tianyun Tang
Potential Business Impact:
Makes complex math problems easier for computers.
Optimal transport (OT) is a powerful tool in mathematics and data science but faces severe computational and statistical challenges in high dimensions. We propose convex relaxation approaches based on marginal and cluster moment relaxations that exploit locality and correlative sparsity in the distributions. These methods approximate high-dimensional couplings using low-order marginals and sparse moment statistics, yielding semidefinite programs that provide lower bounds on the OT cost with greatly reduced complexity. For Gaussian distributions with sparse correlations, we prove reductions in both computational and sample complexity, and experiments show the approach also works well for non-Gaussian cases. In addition, we demonstrate how to extract transport maps from our relaxations, offering a simpler and interpretable alternative to neural networks in generative modeling. Our results suggest that convex relaxations can provide a promising path for dimension reduction in high-dimensional OT.
Similar Papers
Convex relaxation approaches for high dimensional optimal transport
Optimization and Control
Makes complex math problems easier for computers.
A Truncated Newton Method for Optimal Transport
Machine Learning (CS)
Makes computers solve hard math problems faster.
Scalable Approximate Algorithms for Optimal Transport Linear Models
Machine Learning (Stat)
Helps computers learn from data faster and better.