SynJax: Structured Probability Distributions for JAX

TL;DR

SynJax introduces a library that enables efficient, vectorized inference for structured probability distributions in deep learning, facilitating large-scale models that explicitly incorporate data structures like trees and segmentations.

Contribution

It provides the first efficient vectorized implementation of inference algorithms for various structured distributions in JAX, bridging a gap in deep learning software capabilities.

Findings

Enables large-scale structured models in JAX.

Improves efficiency of inference algorithms for structured data.

Facilitates modeling of complex data structures in deep learning.

Abstract

The development of deep learning software libraries enabled significant progress in the field by allowing users to focus on modeling, while letting the library to take care of the tedious and time-consuming task of optimizing execution for modern hardware accelerators. However, this has benefited only particular types of deep learning models, such as Transformers, whose primitives map easily to the vectorized computation. The models that explicitly account for structured objects, such as trees and segmentations, did not benefit equally because they require custom algorithms that are difficult to implement in a vectorized form. SynJax directly addresses this problem by providing an efficient vectorized implementation of inference algorithms for structured distributions covering alignment, tagging, segmentation, constituency trees and spanning trees. This is done by exploiting the connection between algorithms for automatic differentiation and probabilistic inference. With SynJax we can build large-scale differentiable models that explicitly model structure in the data. The code is available at https://github.com/google-deepmind/synjax