Compiler-Level Matrix Multiplication Optimization for Deep Learning

TL;DR

This paper introduces two novel compiler algorithms for optimizing GEMM operations in deep learning, achieving significant performance improvements and faster search times compared to existing methods.

Contribution

It presents two new algorithms based on heuristic search and reinforcement learning for GEMM optimization, outperforming current state-of-the-art techniques.

Findings

Achieved 24% and 40% savings in GEMM computation time.

Explored only 0.1% of the search space.

Demonstrated potential for broader operator-level optimizations.

Abstract

An important linear algebra routine, GEneral Matrix Multiplication (GEMM), is a fundamental operator in deep learning. Compilers need to translate these routines into low-level code optimized for specific hardware. Compiler-level optimization of GEMM has significant performance impact on training and executing deep learning models. However, most deep learning frameworks rely on hardware-specific operator libraries in which GEMM optimization has been mostly achieved by manual tuning, which restricts the performance on different target hardware. In this paper, we propose two novel algorithms for GEMM optimization based on the TVM framework, a lightweight Greedy Best First Search (G-BFS) method based on heuristic search, and a Neighborhood Actor Advantage Critic (N-A2C) method based on reinforcement learning. Experimental results show significant performance improvement of the proposed methods, in both the optimality of the solution and the cost of search in terms of time and fraction of the search space explored. Specifically, the proposed methods achieve 24% and 40% savings in GEMM computation time over state-of-the-art XGBoost and RNN methods, respectively, while exploring only 0.1% of the search space. The proposed approaches have potential to be applied to other operator-level optimizations.