Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

The Sinkhorn algorithm, parabolic optimal transport and geometric Monge-Amp\`ere equations

About

We show that the discrete Sinkhorn algorithm - as applied in the setting of Optimal Transport on a compact manifold - converges to the solution of a fully non-linear parabolic PDE of Monge-Ampere type, in a large-scale limit. The latter evolution equation has previously appeared in different contexts (e.g. on the torus it can be be identified with the Ricci flow). This leads to algorithmic approximations of the potential of the Optimal Transport map, as well as the Optimal Transport distance, with explicit bounds on the arithmetic complexity of the construction and the approximation errors. As applications we obtain explicit schemes of nearly linear complexity, at each iteration, for optimal transport on the torus and the two-sphere, as well as the far-field antenna problem. Connections to Quasi-Monte Carlo methods are exploited.

Robert J. Berman• 2017

Related benchmarks

TaskDatasetResultRank
Domain AdaptationCaltech-office A C (test)
Acc (fi1)91
6
Optimal TransportQuad (Quadratic potential synthetic data)
Epsilon Mu (f_i0)0.047
4
Domain AdaptationCaltech-office A/D (test)
Acc (fi1)87
3
Domain AdaptationCaltech-office C/D (test)
Accuracy (fi1)90
3
Optimal Transport map estimationLog-Sum-Exp
Error e_mu(f_theta_i1)0.006
3
Domain AdaptationCaltech-office A/W (test)
Acc (fi1)78
3
Domain AdaptationCaltech-office C/W (test)
Accuracy (fi1)82
3
Optimal Transport map estimationTensorised
OT Error e_mu(f_theta_i1)0.059
3
Showing 8 of 8 rows

Other info

Follow for update