Presentation
Explainable Fuzzy Neural Network with Multi-Fidelity Reinforcement Learning for Micro-Architecture Design Space Exploration
DescriptionWith 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 the interpretability and controllability of DSE results.
Furthermore, to improve efficiency, we introduce a multi-fidelity reinforcement learning approach, which primarily conducts exploration using inexpensive but imprecise data, thereby substantially diminishing the reliance on costly data.
Experimental results show that our method achieved excellent results with a very limited sample budget and successfully surpasses the current state-of-the-art.
To address this limitation, we propose utilizing Fuzzy Neural Networks to induce and summarize knowledge and insights from the DSE process, enhancing the interpretability and controllability of DSE results.
Furthermore, to improve efficiency, we introduce a multi-fidelity reinforcement learning approach, which primarily conducts exploration using inexpensive but imprecise data, thereby substantially diminishing the reliance on costly data.
Experimental results show that our method achieved excellent results with a very limited sample budget and successfully surpasses the current state-of-the-art.
Event Type
Research Manuscript
TimeWednesday, June 263:30pm - 3:45pm PDT
Location3008, 3rd Floor
EDA
Design Methodologies for System-on-Chip and 3D/2.5D System-in-Package