Optimizing Tensor Train Decomposition in DNNs for RISC-V Architectures Using Design Space Exploration and Compiler Optimizations

TL;DR

This paper presents a comprehensive methodology combining design space exploration and compiler optimizations to efficiently deploy Tensor Train Decomposed DNN layers on RISC-V architectures, significantly improving inference speed.

Contribution

It introduces an end-to-end exploration approach and a specialized tool for optimizing low-rank decompositions of fully connected layers on RISC-V, reducing inference time.

Findings

Tensor Train Decomposition layers run 3x faster than IREE

Decomposed layers are 8x faster than Pluto on the same model

Proposed method effectively optimizes DNN deployment on RISC-V edge devices

Abstract

Deep neural networks (DNNs) have become indispensable in many real-life applications like natural language processing, and autonomous systems. However, deploying DNNs on resource-constrained devices, e.g., in RISC-V platforms, remains challenging due to the high computational and memory demands of fully connected (FC) layers, which dominate resource consumption. Low-rank factorization (LRF) offers an effective approach to compressing FC layers, but the vast design space of LRF solutions involves complex trade-offs among FLOPs, memory size, inference time, and accuracy, making the LRF process complex and time-consuming. This paper introduces an end-to-end LRF design space exploration methodology and a specialized design tool for optimizing FC layers on RISC-V processors. Using Tensor Train Decomposition (TTD) offered by TensorFlow T3F library, the proposed work prunes the LRF design space by excluding first, inefficient decomposition shapes and second, solutions with poor inference performance on RISC-V architectures. Compiler optimizations are then applied to enhance custom T3F layer performance, minimizing inference time and boosting computational efficiency. On average, our TT-decomposed layers run 3x faster than IREE and 8x faster than Pluto on the same compressed model. This work provides an efficient solution for deploying DNNs on edge and embedded devices powered by RISC-V architectures.