← Search

Aleksandar Dimitriev

1 accepted papers

2021

ARMS: Antithetic-REINFORCE-Multi-Sample Gradient for Binary Variables

ICML 2021spotlight

Estimating the gradients for binary variables is a task that arises frequently in various domains, such as training discrete latent variable models. What has been commonly used is a REINFORCE based Monte Carlo estimation method that uses either independent samples or pairs of negatively correlated s…