Google Inc.

Krzysztof Stanislaw Maziarz

Luciano Sbaiz

Jesse Berent

Andrea Gesmundo

Effrosyni Kokiopoulou

Gábor Bartók

Systems and methods for flexible parameter sharing for multi-task learning are provided. A training method can include obtaining a test input, selecting a particular task from one or more tasks, and training a multi-task machine-learned model for the particular task by performing a forward pass using the test input and one or more connection probability matrices to generate a sample distribution of test outputs, training the components of the machine-learned model based at least in part on the sample distribution, and performing a backwards pass to train a connection probability matrix of the multi-task machine-learned model using a straight-through Gumbel-softmax approximation.

Growing community of inventors

Cambridge, United Kingdom

Krzysztof Stanislaw Maziarz