4 papers
Well-Posed KL-Regularized Control via Wasserstein and Kalman-Wasserstein KL Divergences
Viktor Stein, Adwait Datar, Nihat Ay
Kullback-Leibler (KL) divergence regularization is widely used in reinforcement learning, but it becomes infinite under support mismatch and can degenerate in low-noise regimes. Us…
Wasserstein KL-divergence for Gaussian distributions
Adwait Datar, Nihat Ay
We introduce a new version of the KL-divergence for Gaussian distributions which is based on Wasserstein geometry and referred to as WKL-divergence. We show that this version is co…
On the Natural Gradient of the Evidence Lower Bound
Nihat Ay, Jesse van Oostrum, Adwait Datar
This article studies the Fisher-Rao gradient, also referred to as the natural gradient, of the evidence lower bound (ELBO) which plays a central role in generative machine learning…
Convergence Properties of Natural Gradient Descent for Minimizing KL Divergence
Adwait Datar, Nihat Ay
The Kullback-Leibler (KL) divergence plays a central role in probabilistic machine learning, where it commonly serves as the canonical loss function. Optimization in such settings…