"Gen" includes a number of novel language constructs such as a generative function interface to encapsulate probabilistic models, combinators to create new generative functions from existing ones and an inference library providing high-level inference algorithms.
In the study published in the journal PLDI 2019, researchers from the Massachusetts Institute of Technology demonstrated the probabilistic programming system that aims to be both expressive at the modelling level and efficient at the algorithmic level.
"Gen" has already showed better performance than existing probabilistic programming systems for a number of different problems such as tracking objects in space, estimating 3D body pose from a depth image, and inferring the structure of a time series, researchers said.
Based on Julia - a language specialised in numerical analysis and which aims to allow users to express models and create inference algorithms using high-level programming constructs, "Gen" models can be expressed in a number of different ways, each striking a different flexibility/efficiency trade-off. "Gen" provides a built-in modelling language that extends Julia's syntax for function definition.
"Gen" models are black boxes called generative functions (GF), that provide an interface (GFI), exposing capabilities required by inference, researchers said.
The generative function approach is key to making "Gen" suitable for application to a wide range of problems and enables it to use models created in TensorFlow as algorithms written in a programming language or as a result of simulations.
( With inputs from IANS )