Their tool automates the creation of new GPU Kernels that are 1.1x to 3.2x faster than expert-optimized code in tested computations, including 2D convolutions (

Their tool automates the creation of new GPU Kernels that are 1.1x to 3.2x faster than expert-optimized code in tested computations, including 2D convolutions (which are widely used in deep learning models). This tool thus has the potential to boost GPU performance for many computationally intensive applications, including AI applications: "We develop Swizzle Inventor to help programmers implement swizzle programs, by writing program sketches that omit swizzles and delegating their creation to an automatic synthesizer. Our synthesis algorithm scales to real-world programs, allowing us to invent new GPU kernels for stencil computations, matrix transposition, and a finite field multiplication algorithm (used in cryptographic applications). The synthesized 2D convolution and finite-field multiplication kernels are on average 1.5-3.2x and 1.1-1.7x faster, respectively, than expert-optimized CUDA kernels."