Researchers have developed a novel system named “Gen” that can be used for Artificial Intelligence applications such as computer vision, robotics, and statistics without having to deal with equations or manually writing high-performance codes.
“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 study published in 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.
