Explainable Fuzzy Neural Network with Multi-Fidelity Reinforcement Learning for Micro-Architecture Design Space Exploration

TL;DR

This paper introduces an explainable fuzzy neural network combined with multi-fidelity reinforcement learning to improve micro-architecture design space exploration, making the process more interpretable and efficient with limited data.

Contribution

It presents a novel DSE framework that enhances interpretability using fuzzy neural networks and reduces data costs through multi-fidelity reinforcement learning.

Findings

Achieves high-quality results with limited samples

Surpasses current state-of-the-art in DSE performance

Provides an open-source framework for micro-architecture exploration

Abstract

With the continuous advancement of processors, modern micro-architecture designs have become increasingly complex. The vast design space presents significant challenges for human designers, making design space exploration (DSE) algorithms a significant tool for $\mu$-arch design. In recent years, efforts have been made in the development of DSE algorithms, and promising results have been achieved. However, the existing DSE algorithms, e.g., Bayesian Optimization and ensemble learning, suffer from poor interpretability, hindering designers' understanding of the decision-making process. To address this limitation, we propose utilizing Fuzzy Neural Networks to induce and summarize knowledge and insights from the DSE process, enhancing interpretability and controllability. Furthermore, to improve efficiency, we introduce a multi-fidelity reinforcement learning approach, which primarily conducts exploration using cheap but less precise data, thereby substantially diminishing the reliance on costly data. Experimental results show that our method achieves excellent results with a very limited sample budget and successfully surpasses the current state-of-the-art. Our DSE framework is open-sourced and available at https://github.com/fanhanwei/FNN\_MFRL\_ArchDSE/\ .