URL details: dfdazac.github.io/sinkhorn.html
URL title:
Approximating Wasserstein distances with PyTorch - Daniel Daza
URL description:
Many problems in machine learning deal with the idea of making two probability distributions to be as close as possible. In the simpler case where we only have observed variables $\mathbf{x}$ (say, images of cats) coming from an unknown distribution $p(\mathbf{x})$, we’d like to find a model $q(\mathbf{x}\vert\theta)$ (like a neural network) that is a good approximation of $p(\mathbf{x})$. It can be shown1 that minimizing $\text{KL}(p\Vert q)$ is equivalent to minimizing the negative log-likelihood, which i
URL last crawled:
2022-06-30
URL speed:
0.921 MB/s,
downloaded in 0.050 seconds
We found no external links pointing to this url.