Deep Neural Networks (DNNs) have gained popularity in the realm of deep learning due to their learnable parameters that are essential for training and inference phases. Nonetheless, deploying these models on mobile and edge devices, which often have constrained hardware resources and power budgets, poses a considerable challenge.

Achieving real-time performance and energy efficiency becomes imperative, necessitating the compression of DNN models. This research paper introduces a fixed partition compaction method that leverages consecutive zeros and non-zero weights/parameters within sparse DNN models.

The paper resulted from the collective work of our MAGIC cluster team at IIT Palakkad and was released by the Association of Computing Machinery (ACM).