Google Research

Training a Subsampling Mechanism in Expectation

  • Colin Raffel
  • Dieterich Lawson
ICLR Workshop (2017)


We describe a mechanism for subsampling sequences and show how to compute its expected output so that it can be trained with standard backpropagation. We test this approach on a simple toy problem and discuss its shortcomings.

Research Areas

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work