Program Generation for Small-Scale Linear Algebra

Many performance-critical computations in communication, control, multimedia processing, machine learning, or graphics fall into the domain of linear algebra. Existing optimized libraries for linear algebra are usually optimized for large scale computation and for uses in scientific computing. For small scale computations in other domains, they are often suboptimal. In this talk I present our work on generating optimized linear algebra code directly from a mathematical description using techniques developed in Spiral: layers of domain-specific languages to express the mathematics and the use of rewriting systems to reshape the computation at a high level of abstraction to overcome known compiler limitations. (This is the thesis work of Daniele Spampinato.)