The Husky Programming Language for Efficient and Strongly Interpretable AI

We invent a new programming language called Husky (https://github.com/xiyuzhai-husky-lang/husky) for efficient and strongly interpretable AI that can be fundamentally different from deep learning and traditional models. It’s a long term project involving multiple disciplines, including machine learning theory, geometry, programming language, visualization, system/architecture. We expect to create a model for ImageNet in the next 16 months that is as accurate as deep learning but strongly interpretable and much more efficient. We’ll cover theoretical justification why we believe it will work.