A more open efficient future for AI development and data science with an introduction to Julia
Alan S. Edelman · 2017
We propose a more open, efficient, expressive, and ergonomic future for AI development, machine learning, and data science based on the Julia programming language. Our thesis is that the current tapestry of high level codes with library calls creates programmer indirections that can work well for the “one off”, but can slow general progress. We provide examples from Machine Learning, Automatic Differentiation, and Data Handling Technologies.