# Closed-form analytical solutions to Optimal Transport/Wasserstein distance

Mathematics Asked by random_shape on January 13, 2021

Kuang and Tabak (2017) mentions that:

"closed-form solutions of the multidimensional optimal transport problems are relatively rare, a number of numerical algorithms have been proposed."

I’m wondering if there are some resources (lecture notes, papers, etc.) that collect/contain known solutions to optimal transport and/or Wasserstein distance between two distributions in dimensions greater than 1. For example, let $$mathcal{N_1}(mu_1, Sigma_1)$$ and $$mathcal{N_2}(mu_2, Sigma_2)$$ denote two Gaussian distributions with different means and covariances matrices. Then the optimal transport map between them is:

$$x longrightarrow mu_2 + A( x – mu_1 )$$ where $$A = Sigma_1^{- 1/2} (Sigma_1^{1/2} Sigma_2 Sigma_1^{1/2}) Sigma_1^{- 1/2}$$. And so the Wasserstein 2 distance is

$$W_2 ( mathcal{N_1}(mu_1, Sigma_1), mathcal{N_2}(mu_2, Sigma_2) ) = || mu_1 – mu_2 ||^2_2 + mathrm{Tr}( Sigma_1 + Sigma_2 – 2( Sigma_1^{1/2} Sigma_2 Sigma_1^{1/2} )^{1/2} )$$ where $$mathrm{Tr}$$ is the trace operator.

It will be nice to know more worked out examples of optimal transport, such as uniform distributions between different geometric objects, e.g. concentric and overlapping balls, between rectangles, etc.

Optimal transport (OT) problems admit closed-form analytical solutions in a very few notable cases, e.g. in 1D or between Gaussians. Below I cite articles providing analytical solutions for the 1-dimensional case only (does 1D mean univariate?)

Formula 3 in the following gives a closed-form analytical solution for Wasserstein distance in the case of 1-D probability distributions, but a source for the formula isn't given and I wonder how to convert it to a discretized linear programming model:

1. Kolouri et al (2019) "Generalized Sliced Wasserstein Distances" https://arxiv.org/pdf/1902.00434.pdf

Formula 9 in the following also gives a closed-form solution:

1. Kolouri et al (2019) "Sliced-Wasserstein Auto-encoders" https://openreview.net/pdf?id=H1xaJn05FQ

Formula 7 in the article below does as well:

1. Kolouri et al (2017) "Optimal Mass Transport: Signal-processing and machine learning applications" https://www.math.ucdavis.edu/~saito/data/acha.read.s19/kolouri-etal_optimal-mass-transport.pdf

Answered by develarist on January 13, 2021

Although a bit old, this is indeed a good question. Here is my bit on the matter:

1. Regarding Gaussian Mixture Models: A Wasserstein-type distance in the space of Gaussian Mixture Models, Julie Delon and Agnes Desolneux, https://arxiv.org/pdf/1907.05254.pdf

2. Using the 2-Wasserstein metric, Mallasto and Feragen geometrize the space of Gaussian processes with $$L_2$$ mean and covariance functions over compact index spaces: Learning from uncertain curves: The 2-Wasserstein metric for Gaussian processes, Anton Mallasto, Aasa Feragen https://papers.nips.cc/paper/7149-learning-from-uncertain-curves-the-2-wasserstein-metric-for-gaussian-processes.pdf

3. Wasserstein space of elliptical distributions are characterized by Muzellec and Cuturi. Authors show that for elliptical probability distributions, Wasserstein distance can be computed via a simple Riemannian descent procedure: Generalizing Point Embeddings using the Wasserstein Space of Elliptical Distributions, Boris Muzellec and Marco Cuturi https://arxiv.org/pdf/1805.07594.pdf (Not closed form)

4. Tree metrics as ground metrics yield negative definite OT metrics that can be computed in a closed form. Sliced-Wasserstein distance is then a particular (special) case (the tree is a chain): Tree-Sliced Variants of Wasserstein Distances, Tam Le, Makoto Yamada, Kenji Fukumizu, Marco Cuturi https://arxiv.org/pdf/1902.00342.pdf

5. Sinkhorn distances/divergences (Cuturi, 2013) are now treated as new forms of distances (e.g. not approximations to $$mathcal{W}_2^2$$) (Genevay et al, 2019). Recently, this entropy regularized optimal transport distance is found to admit a closed form for Gaussian measures: Janati et al (2020). This fascinating finding also extends to the unbalanced case.

I would be happy to keep this list up to date and evolving.

Answered by Tolga Birdal on January 13, 2021

## Related Questions

### Invariant $SU(3)$ subgroup for ${bf 8}$ in ${bf 3}^* otimes {bf 3} ={bf 1} oplus {bf 8}$

1  Asked on December 3, 2021 by annie-marie-cur

### Proof that a continuous function with continuous right derivatives is differentiable.

1  Asked on December 3, 2021

### Finding the volume when a parabola is rotated about the line $y = 4$.

1  Asked on December 3, 2021

### Differential equation, modulus signs in solution?

2  Asked on December 3, 2021 by refnom95

### Consider the sequence where $a_1>0$, $ka_n>a_{n+1}$ and $0<k<1$. Can we say it converges?

1  Asked on December 3, 2021 by oek-cafu

### In a Reflexive banach space, given a closed convex set $C$ and some point $y$, there is a point in $C$, of minimal distance to $y$

2  Asked on December 3, 2021

### Ball / Urn question with a twist

1  Asked on December 3, 2021 by user109387

### How to prove $phi'(t)1_{Omega_t}(w)$ is measurable?

1  Asked on December 3, 2021 by czzzzzzz

### Using characteristic functions to determine distribution of sum of independent normal random variables.

0  Asked on December 3, 2021 by jkeg

### Radioactivity formula using differential equations?

3  Asked on December 3, 2021 by mitali-mittal

### What is the algebraic interpretation of a contracted product?

0  Asked on December 3, 2021 by james-steele

### $f:[0,1]rightarrow[0,1]$, measurable, and $int_{[0,1]}f(x)dx=yimplies m{x:f(x)>frac{y}{2}}geqfrac{y}{2}$.

1  Asked on December 3, 2021

### Evaluating an integral with a division by $0$ issue

3  Asked on December 1, 2021

### Evaluating $sumlimits_{i=lceil frac{n}{2}rceil}^inftybinom{2i}{n}frac{1}{2^i}$

1  Asked on December 1, 2021 by maxim-enis

### Lie group theory’s connection to fractional calculus?

0  Asked on December 1, 2021

### Solution to autonomous differential equation with locally lipscitz function

1  Asked on December 1, 2021

### Let $A,Bin M_n (mathbb R), lambdain sigma(B), alpha in sigma (B)$ and  be inner product show $=lambda||x||^2+alpha$

0  Asked on December 1, 2021

### Is an automorphism a function or a group?

2  Asked on December 1, 2021 by josh-charleston

### Fundamental Theorem of Projective Geometry for Finite Fields

0  Asked on December 1, 2021 by am2000

### Prove a metric space is totally bounded

2  Asked on December 1, 2021