Introduction to TensorFlow Lattice

Опубликовано 11 октября 2017, 17:04

Check out Gitbub Repo for instructions and examples: goo.gl/LbVVPh
Also take a look at the Deep Lattice Networks publication for mathematical details: goo.gl/REE6cb

Maya Gupta, Seungil You and Jan Pfeifer introduce TensorFlow Lattice. TensorFlow Lattice is an open source project that provides a set of prebuilt TensorFlow Estimators that are easy to use, and TensorFlow operators to build your own lattice models. Lattices are multi-dimensional interpolated look-up tables that can approximate any input-output relationships in your data. Furthermore, TensorFlow Lattice provides a way to impose a monotonic input-output relationship (an input goes up, then the output goes up) and smooth regularizers including Laplacian, Torsion to help your model generalize better for unseen feature space.

Subscribe to the Google Developers Channel: goo.gl/mQyv5L