Coarse Grained Reconfigurable Array

Coarse-grained reconfigurable arrays (CGRAs) are adaptable hardware architectures designed for efficient execution of various computational tasks, particularly in resource-constrained environments like edge AI devices. Current research focuses on optimizing application mapping onto CGRAs, employing techniques such as reinforcement learning with graph neural networks to automate the complex process of assigning operations to processing elements and scheduling their execution. This research aims to improve the performance and energy efficiency of CGRAs, enabling their wider adoption in applications requiring high throughput and low power consumption.